opencv-lib - openCV library for Renesas RZ/A

Renesas » Code » opencv-lib

openCV library for Renesas RZ/A

include/opencv2/core.hpp@0:0e0631af0305, 2021-01-29 (annotated)

Committer:: RyoheiHagimoto
Date:: Fri Jan 29 04:53:38 2021 +0000
Revision:: 0:0e0631af0305

copied from https://github.com/d-kato/opencv-lib.

Who changed what in which revision?

User	Revision	Line number	New contents of line
RyoheiHagimoto	0:0e0631af0305	1	/*M///////////////////////////////////////////////////////////////////////////////////////
RyoheiHagimoto	0:0e0631af0305	2	//
RyoheiHagimoto	0:0e0631af0305	3	// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
RyoheiHagimoto	0:0e0631af0305	4	//
RyoheiHagimoto	0:0e0631af0305	5	// By downloading, copying, installing or using the software you agree to this license.
RyoheiHagimoto	0:0e0631af0305	6	// If you do not agree to this license, do not download, install,
RyoheiHagimoto	0:0e0631af0305	7	// copy or use the software.
RyoheiHagimoto	0:0e0631af0305	8	//
RyoheiHagimoto	0:0e0631af0305	9	//
RyoheiHagimoto	0:0e0631af0305	10	// License Agreement
RyoheiHagimoto	0:0e0631af0305	11	// For Open Source Computer Vision Library
RyoheiHagimoto	0:0e0631af0305	12	//
RyoheiHagimoto	0:0e0631af0305	13	// Copyright (C) 2000-2015, Intel Corporation, all rights reserved.
RyoheiHagimoto	0:0e0631af0305	14	// Copyright (C) 2009-2011, Willow Garage Inc., all rights reserved.
RyoheiHagimoto	0:0e0631af0305	15	// Copyright (C) 2015, OpenCV Foundation, all rights reserved.
RyoheiHagimoto	0:0e0631af0305	16	// Copyright (C) 2015, Itseez Inc., all rights reserved.
RyoheiHagimoto	0:0e0631af0305	17	// Third party copyrights are property of their respective owners.
RyoheiHagimoto	0:0e0631af0305	18	//
RyoheiHagimoto	0:0e0631af0305	19	// Redistribution and use in source and binary forms, with or without modification,
RyoheiHagimoto	0:0e0631af0305	20	// are permitted provided that the following conditions are met:
RyoheiHagimoto	0:0e0631af0305	21	//
RyoheiHagimoto	0:0e0631af0305	22	// * Redistribution's of source code must retain the above copyright notice,
RyoheiHagimoto	0:0e0631af0305	23	// this list of conditions and the following disclaimer.
RyoheiHagimoto	0:0e0631af0305	24	//
RyoheiHagimoto	0:0e0631af0305	25	// * Redistribution's in binary form must reproduce the above copyright notice,
RyoheiHagimoto	0:0e0631af0305	26	// this list of conditions and the following disclaimer in the documentation
RyoheiHagimoto	0:0e0631af0305	27	// and/or other materials provided with the distribution.
RyoheiHagimoto	0:0e0631af0305	28	//
RyoheiHagimoto	0:0e0631af0305	29	// * The name of the copyright holders may not be used to endorse or promote products
RyoheiHagimoto	0:0e0631af0305	30	// derived from this software without specific prior written permission.
RyoheiHagimoto	0:0e0631af0305	31	//
RyoheiHagimoto	0:0e0631af0305	32	// This software is provided by the copyright holders and contributors "as is" and
RyoheiHagimoto	0:0e0631af0305	33	// any express or implied warranties, including, but not limited to, the implied
RyoheiHagimoto	0:0e0631af0305	34	// warranties of merchantability and fitness for a particular purpose are disclaimed.
RyoheiHagimoto	0:0e0631af0305	35	// In no event shall the Intel Corporation or contributors be liable for any direct,
RyoheiHagimoto	0:0e0631af0305	36	// indirect, incidental, special, exemplary, or consequential damages
RyoheiHagimoto	0:0e0631af0305	37	// (including, but not limited to, procurement of substitute goods or services;
RyoheiHagimoto	0:0e0631af0305	38	// loss of use, data, or profits; or business interruption) however caused
RyoheiHagimoto	0:0e0631af0305	39	// and on any theory of liability, whether in contract, strict liability,
RyoheiHagimoto	0:0e0631af0305	40	// or tort (including negligence or otherwise) arising in any way out of
RyoheiHagimoto	0:0e0631af0305	41	// the use of this software, even if advised of the possibility of such damage.
RyoheiHagimoto	0:0e0631af0305	42	//
RyoheiHagimoto	0:0e0631af0305	43	//M*/
RyoheiHagimoto	0:0e0631af0305	44
RyoheiHagimoto	0:0e0631af0305	45	#ifndef OPENCV_CORE_HPP
RyoheiHagimoto	0:0e0631af0305	46	#define OPENCV_CORE_HPP
RyoheiHagimoto	0:0e0631af0305	47
RyoheiHagimoto	0:0e0631af0305	48	#ifndef __cplusplus
RyoheiHagimoto	0:0e0631af0305	49	# error core.hpp header must be compiled as C++
RyoheiHagimoto	0:0e0631af0305	50	#endif
RyoheiHagimoto	0:0e0631af0305	51
RyoheiHagimoto	0:0e0631af0305	52	#include "opencv2/core/cvdef.h"
RyoheiHagimoto	0:0e0631af0305	53	#include "opencv2/core/version.hpp"
RyoheiHagimoto	0:0e0631af0305	54	#include "opencv2/core/base.hpp"
RyoheiHagimoto	0:0e0631af0305	55	#include "opencv2/core/cvstd.hpp"
RyoheiHagimoto	0:0e0631af0305	56	#include "opencv2/core/traits.hpp"
RyoheiHagimoto	0:0e0631af0305	57	#include "opencv2/core/matx.hpp"
RyoheiHagimoto	0:0e0631af0305	58	#include "opencv2/core/types.hpp"
RyoheiHagimoto	0:0e0631af0305	59	#include "opencv2/core/mat.hpp"
RyoheiHagimoto	0:0e0631af0305	60	#include "opencv2/core/persistence.hpp"
RyoheiHagimoto	0:0e0631af0305	61
RyoheiHagimoto	0:0e0631af0305	62	/**
RyoheiHagimoto	0:0e0631af0305	63	@defgroup core Core functionality
RyoheiHagimoto	0:0e0631af0305	64	@{
RyoheiHagimoto	0:0e0631af0305	65	@defgroup core_basic Basic structures
RyoheiHagimoto	0:0e0631af0305	66	@defgroup core_c C structures and operations
RyoheiHagimoto	0:0e0631af0305	67	@{
RyoheiHagimoto	0:0e0631af0305	68	@defgroup core_c_glue Connections with C++
RyoheiHagimoto	0:0e0631af0305	69	@}
RyoheiHagimoto	0:0e0631af0305	70	@defgroup core_array Operations on arrays
RyoheiHagimoto	0:0e0631af0305	71	@defgroup core_xml XML/YAML Persistence
RyoheiHagimoto	0:0e0631af0305	72	@defgroup core_cluster Clustering
RyoheiHagimoto	0:0e0631af0305	73	@defgroup core_utils Utility and system functions and macros
RyoheiHagimoto	0:0e0631af0305	74	@{
RyoheiHagimoto	0:0e0631af0305	75	@defgroup core_utils_sse SSE utilities
RyoheiHagimoto	0:0e0631af0305	76	@defgroup core_utils_neon NEON utilities
RyoheiHagimoto	0:0e0631af0305	77	@}
RyoheiHagimoto	0:0e0631af0305	78	@defgroup core_opengl OpenGL interoperability
RyoheiHagimoto	0:0e0631af0305	79	@defgroup core_ipp Intel IPP Asynchronous C/C++ Converters
RyoheiHagimoto	0:0e0631af0305	80	@defgroup core_optim Optimization Algorithms
RyoheiHagimoto	0:0e0631af0305	81	@defgroup core_directx DirectX interoperability
RyoheiHagimoto	0:0e0631af0305	82	@defgroup core_eigen Eigen support
RyoheiHagimoto	0:0e0631af0305	83	@defgroup core_opencl OpenCL support
RyoheiHagimoto	0:0e0631af0305	84	@defgroup core_va_intel Intel VA-API/OpenCL (CL-VA) interoperability
RyoheiHagimoto	0:0e0631af0305	85	@defgroup core_hal Hardware Acceleration Layer
RyoheiHagimoto	0:0e0631af0305	86	@{
RyoheiHagimoto	0:0e0631af0305	87	@defgroup core_hal_functions Functions
RyoheiHagimoto	0:0e0631af0305	88	@defgroup core_hal_interface Interface
RyoheiHagimoto	0:0e0631af0305	89	@defgroup core_hal_intrin Universal intrinsics
RyoheiHagimoto	0:0e0631af0305	90	@{
RyoheiHagimoto	0:0e0631af0305	91	@defgroup core_hal_intrin_impl Private implementation helpers
RyoheiHagimoto	0:0e0631af0305	92	@}
RyoheiHagimoto	0:0e0631af0305	93	@}
RyoheiHagimoto	0:0e0631af0305	94	@}
RyoheiHagimoto	0:0e0631af0305	95	*/
RyoheiHagimoto	0:0e0631af0305	96
RyoheiHagimoto	0:0e0631af0305	97	namespace cv {
RyoheiHagimoto	0:0e0631af0305	98
RyoheiHagimoto	0:0e0631af0305	99	//! @addtogroup core_utils
RyoheiHagimoto	0:0e0631af0305	100	//! @{
RyoheiHagimoto	0:0e0631af0305	101
RyoheiHagimoto	0:0e0631af0305	102	/*! @brief Class passed to an error.
RyoheiHagimoto	0:0e0631af0305	103
RyoheiHagimoto	0:0e0631af0305	104	This class encapsulates all or almost all necessary
RyoheiHagimoto	0:0e0631af0305	105	information about the error happened in the program. The exception is
RyoheiHagimoto	0:0e0631af0305	106	usually constructed and thrown implicitly via CV_Error and CV_Error_ macros.
RyoheiHagimoto	0:0e0631af0305	107	@see error
RyoheiHagimoto	0:0e0631af0305	108	*/
RyoheiHagimoto	0:0e0631af0305	109	class CV_EXPORTS Exception : public std::exception
RyoheiHagimoto	0:0e0631af0305	110	{
RyoheiHagimoto	0:0e0631af0305	111	public:
RyoheiHagimoto	0:0e0631af0305	112	/*!
RyoheiHagimoto	0:0e0631af0305	113	Default constructor
RyoheiHagimoto	0:0e0631af0305	114	*/
RyoheiHagimoto	0:0e0631af0305	115	Exception();
RyoheiHagimoto	0:0e0631af0305	116	/*!
RyoheiHagimoto	0:0e0631af0305	117	Full constructor. Normally the constuctor is not called explicitly.
RyoheiHagimoto	0:0e0631af0305	118	Instead, the macros CV_Error(), CV_Error_() and CV_Assert() are used.
RyoheiHagimoto	0:0e0631af0305	119	*/
RyoheiHagimoto	0:0e0631af0305	120	Exception(int _code, const String& _err, const String& _func, const String& _file, int _line);
RyoheiHagimoto	0:0e0631af0305	121	virtual ~Exception() throw();
RyoheiHagimoto	0:0e0631af0305	122
RyoheiHagimoto	0:0e0631af0305	123	/*!
RyoheiHagimoto	0:0e0631af0305	124	\return the error description and the context as a text string.
RyoheiHagimoto	0:0e0631af0305	125	*/
RyoheiHagimoto	0:0e0631af0305	126	virtual const char *what() const throw();
RyoheiHagimoto	0:0e0631af0305	127	void formatMessage();
RyoheiHagimoto	0:0e0631af0305	128
RyoheiHagimoto	0:0e0631af0305	129	String msg; ///< the formatted error message
RyoheiHagimoto	0:0e0631af0305	130
RyoheiHagimoto	0:0e0631af0305	131	int code; ///< error code @see CVStatus
RyoheiHagimoto	0:0e0631af0305	132	String err; ///< error description
RyoheiHagimoto	0:0e0631af0305	133	String func; ///< function name. Available only when the compiler supports getting it
RyoheiHagimoto	0:0e0631af0305	134	String file; ///< source file name where the error has occured
RyoheiHagimoto	0:0e0631af0305	135	int line; ///< line number in the source file where the error has occured
RyoheiHagimoto	0:0e0631af0305	136	};
RyoheiHagimoto	0:0e0631af0305	137
RyoheiHagimoto	0:0e0631af0305	138	/*! @brief Signals an error and raises the exception.
RyoheiHagimoto	0:0e0631af0305	139
RyoheiHagimoto	0:0e0631af0305	140	By default the function prints information about the error to stderr,
RyoheiHagimoto	0:0e0631af0305	141	then it either stops if cv::setBreakOnError() had been called before or raises the exception.
RyoheiHagimoto	0:0e0631af0305	142	It is possible to alternate error processing by using cv::redirectError().
RyoheiHagimoto	0:0e0631af0305	143	@param exc the exception raisen.
RyoheiHagimoto	0:0e0631af0305	144	@deprecated drop this version
RyoheiHagimoto	0:0e0631af0305	145	*/
RyoheiHagimoto	0:0e0631af0305	146	CV_EXPORTS void error( const Exception& exc );
RyoheiHagimoto	0:0e0631af0305	147
RyoheiHagimoto	0:0e0631af0305	148	enum SortFlags { SORT_EVERY_ROW = 0, //!< each matrix row is sorted independently
RyoheiHagimoto	0:0e0631af0305	149	SORT_EVERY_COLUMN = 1, //!< each matrix column is sorted
RyoheiHagimoto	0:0e0631af0305	150	//!< independently; this flag and the previous one are
RyoheiHagimoto	0:0e0631af0305	151	//!< mutually exclusive.
RyoheiHagimoto	0:0e0631af0305	152	SORT_ASCENDING = 0, //!< each matrix row is sorted in the ascending
RyoheiHagimoto	0:0e0631af0305	153	//!< order.
RyoheiHagimoto	0:0e0631af0305	154	SORT_DESCENDING = 16 //!< each matrix row is sorted in the
RyoheiHagimoto	0:0e0631af0305	155	//!< descending order; this flag and the previous one are also
RyoheiHagimoto	0:0e0631af0305	156	//!< mutually exclusive.
RyoheiHagimoto	0:0e0631af0305	157	};
RyoheiHagimoto	0:0e0631af0305	158
RyoheiHagimoto	0:0e0631af0305	159	//! @} core_utils
RyoheiHagimoto	0:0e0631af0305	160
RyoheiHagimoto	0:0e0631af0305	161	//! @addtogroup core
RyoheiHagimoto	0:0e0631af0305	162	//! @{
RyoheiHagimoto	0:0e0631af0305	163
RyoheiHagimoto	0:0e0631af0305	164	//! Covariation flags
RyoheiHagimoto	0:0e0631af0305	165	enum CovarFlags {
RyoheiHagimoto	0:0e0631af0305	166	/** The output covariance matrix is calculated as:
RyoheiHagimoto	0:0e0631af0305	167	\f[\texttt{scale} \cdot [ \texttt{vects} [0]- \texttt{mean} , \texttt{vects} [1]- \texttt{mean} ,...]^T \cdot [ \texttt{vects} [0]- \texttt{mean} , \texttt{vects} [1]- \texttt{mean} ,...],\f]
RyoheiHagimoto	0:0e0631af0305	168	The covariance matrix will be nsamples x nsamples. Such an unusual covariance matrix is used
RyoheiHagimoto	0:0e0631af0305	169	for fast PCA of a set of very large vectors (see, for example, the EigenFaces technique for
RyoheiHagimoto	0:0e0631af0305	170	face recognition). Eigenvalues of this "scrambled" matrix match the eigenvalues of the true
RyoheiHagimoto	0:0e0631af0305	171	covariance matrix. The "true" eigenvectors can be easily calculated from the eigenvectors of
RyoheiHagimoto	0:0e0631af0305	172	the "scrambled" covariance matrix. */
RyoheiHagimoto	0:0e0631af0305	173	COVAR_SCRAMBLED = 0,
RyoheiHagimoto	0:0e0631af0305	174	/**The output covariance matrix is calculated as:
RyoheiHagimoto	0:0e0631af0305	175	\f[\texttt{scale} \cdot [ \texttt{vects} [0]- \texttt{mean} , \texttt{vects} [1]- \texttt{mean} ,...] \cdot [ \texttt{vects} [0]- \texttt{mean} , \texttt{vects} [1]- \texttt{mean} ,...]^T,\f]
RyoheiHagimoto	0:0e0631af0305	176	covar will be a square matrix of the same size as the total number of elements in each input
RyoheiHagimoto	0:0e0631af0305	177	vector. One and only one of COVAR_SCRAMBLED and COVAR_NORMAL must be specified.*/
RyoheiHagimoto	0:0e0631af0305	178	COVAR_NORMAL = 1,
RyoheiHagimoto	0:0e0631af0305	179	/** If the flag is specified, the function does not calculate mean from
RyoheiHagimoto	0:0e0631af0305	180	the input vectors but, instead, uses the passed mean vector. This is useful if mean has been
RyoheiHagimoto	0:0e0631af0305	181	pre-calculated or known in advance, or if the covariance matrix is calculated by parts. In
RyoheiHagimoto	0:0e0631af0305	182	this case, mean is not a mean vector of the input sub-set of vectors but rather the mean
RyoheiHagimoto	0:0e0631af0305	183	vector of the whole set.*/
RyoheiHagimoto	0:0e0631af0305	184	COVAR_USE_AVG = 2,
RyoheiHagimoto	0:0e0631af0305	185	/** If the flag is specified, the covariance matrix is scaled. In the
RyoheiHagimoto	0:0e0631af0305	186	"normal" mode, scale is 1./nsamples . In the "scrambled" mode, scale is the reciprocal of the
RyoheiHagimoto	0:0e0631af0305	187	total number of elements in each input vector. By default (if the flag is not specified), the
RyoheiHagimoto	0:0e0631af0305	188	covariance matrix is not scaled ( scale=1 ).*/
RyoheiHagimoto	0:0e0631af0305	189	COVAR_SCALE = 4,
RyoheiHagimoto	0:0e0631af0305	190	/** If the flag is
RyoheiHagimoto	0:0e0631af0305	191	specified, all the input vectors are stored as rows of the samples matrix. mean should be a
RyoheiHagimoto	0:0e0631af0305	192	single-row vector in this case.*/
RyoheiHagimoto	0:0e0631af0305	193	COVAR_ROWS = 8,
RyoheiHagimoto	0:0e0631af0305	194	/** If the flag is
RyoheiHagimoto	0:0e0631af0305	195	specified, all the input vectors are stored as columns of the samples matrix. mean should be a
RyoheiHagimoto	0:0e0631af0305	196	single-column vector in this case.*/
RyoheiHagimoto	0:0e0631af0305	197	COVAR_COLS = 16
RyoheiHagimoto	0:0e0631af0305	198	};
RyoheiHagimoto	0:0e0631af0305	199
RyoheiHagimoto	0:0e0631af0305	200	//! k-Means flags
RyoheiHagimoto	0:0e0631af0305	201	enum KmeansFlags {
RyoheiHagimoto	0:0e0631af0305	202	/** Select random initial centers in each attempt.*/
RyoheiHagimoto	0:0e0631af0305	203	KMEANS_RANDOM_CENTERS = 0,
RyoheiHagimoto	0:0e0631af0305	204	/** Use kmeans++ center initialization by Arthur and Vassilvitskii [Arthur2007].*/
RyoheiHagimoto	0:0e0631af0305	205	KMEANS_PP_CENTERS = 2,
RyoheiHagimoto	0:0e0631af0305	206	/** During the first (and possibly the only) attempt, use the
RyoheiHagimoto	0:0e0631af0305	207	user-supplied labels instead of computing them from the initial centers. For the second and
RyoheiHagimoto	0:0e0631af0305	208	further attempts, use the random or semi-random centers. Use one of KMEANS_\*_CENTERS flag
RyoheiHagimoto	0:0e0631af0305	209	to specify the exact method.*/
RyoheiHagimoto	0:0e0631af0305	210	KMEANS_USE_INITIAL_LABELS = 1
RyoheiHagimoto	0:0e0631af0305	211	};
RyoheiHagimoto	0:0e0631af0305	212
RyoheiHagimoto	0:0e0631af0305	213	//! type of line
RyoheiHagimoto	0:0e0631af0305	214	enum LineTypes {
RyoheiHagimoto	0:0e0631af0305	215	FILLED = -1,
RyoheiHagimoto	0:0e0631af0305	216	LINE_4 = 4, //!< 4-connected line
RyoheiHagimoto	0:0e0631af0305	217	LINE_8 = 8, //!< 8-connected line
RyoheiHagimoto	0:0e0631af0305	218	LINE_AA = 16 //!< antialiased line
RyoheiHagimoto	0:0e0631af0305	219	};
RyoheiHagimoto	0:0e0631af0305	220
RyoheiHagimoto	0:0e0631af0305	221	//! Only a subset of Hershey fonts
RyoheiHagimoto	0:0e0631af0305	222	//! <http://sources.isc.org/utils/misc/hershey-font.txt> are supported
RyoheiHagimoto	0:0e0631af0305	223	enum HersheyFonts {
RyoheiHagimoto	0:0e0631af0305	224	FONT_HERSHEY_SIMPLEX = 0, //!< normal size sans-serif font
RyoheiHagimoto	0:0e0631af0305	225	FONT_HERSHEY_PLAIN = 1, //!< small size sans-serif font
RyoheiHagimoto	0:0e0631af0305	226	FONT_HERSHEY_DUPLEX = 2, //!< normal size sans-serif font (more complex than FONT_HERSHEY_SIMPLEX)
RyoheiHagimoto	0:0e0631af0305	227	FONT_HERSHEY_COMPLEX = 3, //!< normal size serif font
RyoheiHagimoto	0:0e0631af0305	228	FONT_HERSHEY_TRIPLEX = 4, //!< normal size serif font (more complex than FONT_HERSHEY_COMPLEX)
RyoheiHagimoto	0:0e0631af0305	229	FONT_HERSHEY_COMPLEX_SMALL = 5, //!< smaller version of FONT_HERSHEY_COMPLEX
RyoheiHagimoto	0:0e0631af0305	230	FONT_HERSHEY_SCRIPT_SIMPLEX = 6, //!< hand-writing style font
RyoheiHagimoto	0:0e0631af0305	231	FONT_HERSHEY_SCRIPT_COMPLEX = 7, //!< more complex variant of FONT_HERSHEY_SCRIPT_SIMPLEX
RyoheiHagimoto	0:0e0631af0305	232	FONT_ITALIC = 16 //!< flag for italic font
RyoheiHagimoto	0:0e0631af0305	233	};
RyoheiHagimoto	0:0e0631af0305	234
RyoheiHagimoto	0:0e0631af0305	235	enum ReduceTypes { REDUCE_SUM = 0, //!< the output is the sum of all rows/columns of the matrix.
RyoheiHagimoto	0:0e0631af0305	236	REDUCE_AVG = 1, //!< the output is the mean vector of all rows/columns of the matrix.
RyoheiHagimoto	0:0e0631af0305	237	REDUCE_MAX = 2, //!< the output is the maximum (column/row-wise) of all rows/columns of the matrix.
RyoheiHagimoto	0:0e0631af0305	238	REDUCE_MIN = 3 //!< the output is the minimum (column/row-wise) of all rows/columns of the matrix.
RyoheiHagimoto	0:0e0631af0305	239	};
RyoheiHagimoto	0:0e0631af0305	240
RyoheiHagimoto	0:0e0631af0305	241
RyoheiHagimoto	0:0e0631af0305	242	/** @brief Swaps two matrices
RyoheiHagimoto	0:0e0631af0305	243	*/
RyoheiHagimoto	0:0e0631af0305	244	CV_EXPORTS void swap(Mat& a, Mat& b);
RyoheiHagimoto	0:0e0631af0305	245	/** @overload */
RyoheiHagimoto	0:0e0631af0305	246	CV_EXPORTS void swap( UMat& a, UMat& b );
RyoheiHagimoto	0:0e0631af0305	247
RyoheiHagimoto	0:0e0631af0305	248	//! @} core
RyoheiHagimoto	0:0e0631af0305	249
RyoheiHagimoto	0:0e0631af0305	250	//! @addtogroup core_array
RyoheiHagimoto	0:0e0631af0305	251	//! @{
RyoheiHagimoto	0:0e0631af0305	252
RyoheiHagimoto	0:0e0631af0305	253	/** @brief Computes the source location of an extrapolated pixel.
RyoheiHagimoto	0:0e0631af0305	254
RyoheiHagimoto	0:0e0631af0305	255	The function computes and returns the coordinate of a donor pixel corresponding to the specified
RyoheiHagimoto	0:0e0631af0305	256	extrapolated pixel when using the specified extrapolation border mode. For example, if you use
RyoheiHagimoto	0:0e0631af0305	257	cv::BORDER_WRAP mode in the horizontal direction, cv::BORDER_REFLECT_101 in the vertical direction and
RyoheiHagimoto	0:0e0631af0305	258	want to compute value of the "virtual" pixel Point(-5, 100) in a floating-point image img , it
RyoheiHagimoto	0:0e0631af0305	259	looks like:
RyoheiHagimoto	0:0e0631af0305	260	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	261	float val = img.at<float>(borderInterpolate(100, img.rows, cv::BORDER_REFLECT_101),
RyoheiHagimoto	0:0e0631af0305	262	borderInterpolate(-5, img.cols, cv::BORDER_WRAP));
RyoheiHagimoto	0:0e0631af0305	263	@endcode
RyoheiHagimoto	0:0e0631af0305	264	Normally, the function is not called directly. It is used inside filtering functions and also in
RyoheiHagimoto	0:0e0631af0305	265	copyMakeBorder.
RyoheiHagimoto	0:0e0631af0305	266	@param p 0-based coordinate of the extrapolated pixel along one of the axes, likely \<0 or \>= len
RyoheiHagimoto	0:0e0631af0305	267	@param len Length of the array along the corresponding axis.
RyoheiHagimoto	0:0e0631af0305	268	@param borderType Border type, one of the cv::BorderTypes, except for cv::BORDER_TRANSPARENT and
RyoheiHagimoto	0:0e0631af0305	269	cv::BORDER_ISOLATED . When borderType==cv::BORDER_CONSTANT , the function always returns -1, regardless
RyoheiHagimoto	0:0e0631af0305	270	of p and len.
RyoheiHagimoto	0:0e0631af0305	271
RyoheiHagimoto	0:0e0631af0305	272	@sa copyMakeBorder
RyoheiHagimoto	0:0e0631af0305	273	*/
RyoheiHagimoto	0:0e0631af0305	274	CV_EXPORTS_W int borderInterpolate(int p, int len, int borderType);
RyoheiHagimoto	0:0e0631af0305	275
RyoheiHagimoto	0:0e0631af0305	276	/** @brief Forms a border around an image.
RyoheiHagimoto	0:0e0631af0305	277
RyoheiHagimoto	0:0e0631af0305	278	The function copies the source image into the middle of the destination image. The areas to the
RyoheiHagimoto	0:0e0631af0305	279	left, to the right, above and below the copied source image will be filled with extrapolated
RyoheiHagimoto	0:0e0631af0305	280	pixels. This is not what filtering functions based on it do (they extrapolate pixels on-fly), but
RyoheiHagimoto	0:0e0631af0305	281	what other more complex functions, including your own, may do to simplify image boundary handling.
RyoheiHagimoto	0:0e0631af0305	282
RyoheiHagimoto	0:0e0631af0305	283	The function supports the mode when src is already in the middle of dst . In this case, the
RyoheiHagimoto	0:0e0631af0305	284	function does not copy src itself but simply constructs the border, for example:
RyoheiHagimoto	0:0e0631af0305	285
RyoheiHagimoto	0:0e0631af0305	286	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	287	// let border be the same in all directions
RyoheiHagimoto	0:0e0631af0305	288	int border=2;
RyoheiHagimoto	0:0e0631af0305	289	// constructs a larger image to fit both the image and the border
RyoheiHagimoto	0:0e0631af0305	290	Mat gray_buf(rgb.rows + border2, rgb.cols + border2, rgb.depth());
RyoheiHagimoto	0:0e0631af0305	291	// select the middle part of it w/o copying data
RyoheiHagimoto	0:0e0631af0305	292	Mat gray(gray_canvas, Rect(border, border, rgb.cols, rgb.rows));
RyoheiHagimoto	0:0e0631af0305	293	// convert image from RGB to grayscale
RyoheiHagimoto	0:0e0631af0305	294	cvtColor(rgb, gray, COLOR_RGB2GRAY);
RyoheiHagimoto	0:0e0631af0305	295	// form a border in-place
RyoheiHagimoto	0:0e0631af0305	296	copyMakeBorder(gray, gray_buf, border, border,
RyoheiHagimoto	0:0e0631af0305	297	border, border, BORDER_REPLICATE);
RyoheiHagimoto	0:0e0631af0305	298	// now do some custom filtering ...
RyoheiHagimoto	0:0e0631af0305	299	...
RyoheiHagimoto	0:0e0631af0305	300	@endcode
RyoheiHagimoto	0:0e0631af0305	301	@note When the source image is a part (ROI) of a bigger image, the function will try to use the
RyoheiHagimoto	0:0e0631af0305	302	pixels outside of the ROI to form a border. To disable this feature and always do extrapolation, as
RyoheiHagimoto	0:0e0631af0305	303	if src was not a ROI, use borderType \| BORDER_ISOLATED.
RyoheiHagimoto	0:0e0631af0305	304
RyoheiHagimoto	0:0e0631af0305	305	@param src Source image.
RyoheiHagimoto	0:0e0631af0305	306	@param dst Destination image of the same type as src and the size Size(src.cols+left+right,
RyoheiHagimoto	0:0e0631af0305	307	src.rows+top+bottom) .
RyoheiHagimoto	0:0e0631af0305	308	@param top
RyoheiHagimoto	0:0e0631af0305	309	@param bottom
RyoheiHagimoto	0:0e0631af0305	310	@param left
RyoheiHagimoto	0:0e0631af0305	311	@param right Parameter specifying how many pixels in each direction from the source image rectangle
RyoheiHagimoto	0:0e0631af0305	312	to extrapolate. For example, top=1, bottom=1, left=1, right=1 mean that 1 pixel-wide border needs
RyoheiHagimoto	0:0e0631af0305	313	to be built.
RyoheiHagimoto	0:0e0631af0305	314	@param borderType Border type. See borderInterpolate for details.
RyoheiHagimoto	0:0e0631af0305	315	@param value Border value if borderType==BORDER_CONSTANT .
RyoheiHagimoto	0:0e0631af0305	316
RyoheiHagimoto	0:0e0631af0305	317	@sa borderInterpolate
RyoheiHagimoto	0:0e0631af0305	318	*/
RyoheiHagimoto	0:0e0631af0305	319	CV_EXPORTS_W void copyMakeBorder(InputArray src, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	320	int top, int bottom, int left, int right,
RyoheiHagimoto	0:0e0631af0305	321	int borderType, const Scalar& value = Scalar() );
RyoheiHagimoto	0:0e0631af0305	322
RyoheiHagimoto	0:0e0631af0305	323	/** @brief Calculates the per-element sum of two arrays or an array and a scalar.
RyoheiHagimoto	0:0e0631af0305	324
RyoheiHagimoto	0:0e0631af0305	325	The function add calculates:
RyoheiHagimoto	0:0e0631af0305	326	- Sum of two arrays when both input arrays have the same size and the same number of channels:
RyoheiHagimoto	0:0e0631af0305	327	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	328	- Sum of an array and a scalar when src2 is constructed from Scalar or has the same number of
RyoheiHagimoto	0:0e0631af0305	329	elements as `src1.channels()`:
RyoheiHagimoto	0:0e0631af0305	330	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	331	- Sum of a scalar and an array when src1 is constructed from Scalar or has the same number of
RyoheiHagimoto	0:0e0631af0305	332	elements as `src2.channels()`:
RyoheiHagimoto	0:0e0631af0305	333	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} + \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	334	where `I` is a multi-dimensional index of array elements. In case of multi-channel arrays, each
RyoheiHagimoto	0:0e0631af0305	335	channel is processed independently.
RyoheiHagimoto	0:0e0631af0305	336
RyoheiHagimoto	0:0e0631af0305	337	The first function in the list above can be replaced with matrix expressions:
RyoheiHagimoto	0:0e0631af0305	338	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	339	dst = src1 + src2;
RyoheiHagimoto	0:0e0631af0305	340	dst += src1; // equivalent to add(dst, src1, dst);
RyoheiHagimoto	0:0e0631af0305	341	@endcode
RyoheiHagimoto	0:0e0631af0305	342	The input arrays and the output array can all have the same or different depths. For example, you
RyoheiHagimoto	0:0e0631af0305	343	can add a 16-bit unsigned array to a 8-bit signed array and store the sum as a 32-bit
RyoheiHagimoto	0:0e0631af0305	344	floating-point array. Depth of the output array is determined by the dtype parameter. In the second
RyoheiHagimoto	0:0e0631af0305	345	and third cases above, as well as in the first case, when src1.depth() == src2.depth(), dtype can
RyoheiHagimoto	0:0e0631af0305	346	be set to the default -1. In this case, the output array will have the same depth as the input
RyoheiHagimoto	0:0e0631af0305	347	array, be it src1, src2 or both.
RyoheiHagimoto	0:0e0631af0305	348	@note Saturation is not applied when the output array has the depth CV_32S. You may even get
RyoheiHagimoto	0:0e0631af0305	349	result of an incorrect sign in the case of overflow.
RyoheiHagimoto	0:0e0631af0305	350	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	351	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	352	@param dst output array that has the same size and number of channels as the input array(s); the
RyoheiHagimoto	0:0e0631af0305	353	depth is defined by dtype or src1/src2.
RyoheiHagimoto	0:0e0631af0305	354	@param mask optional operation mask - 8-bit single channel array, that specifies elements of the
RyoheiHagimoto	0:0e0631af0305	355	output array to be changed.
RyoheiHagimoto	0:0e0631af0305	356	@param dtype optional depth of the output array (see the discussion below).
RyoheiHagimoto	0:0e0631af0305	357	@sa subtract, addWeighted, scaleAdd, Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	358	*/
RyoheiHagimoto	0:0e0631af0305	359	CV_EXPORTS_W void add(InputArray src1, InputArray src2, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	360	InputArray mask = noArray(), int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	361
RyoheiHagimoto	0:0e0631af0305	362	/** @brief Calculates the per-element difference between two arrays or array and a scalar.
RyoheiHagimoto	0:0e0631af0305	363
RyoheiHagimoto	0:0e0631af0305	364	The function subtract calculates:
RyoheiHagimoto	0:0e0631af0305	365	- Difference between two arrays, when both input arrays have the same size and the same number of
RyoheiHagimoto	0:0e0631af0305	366	channels:
RyoheiHagimoto	0:0e0631af0305	367	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	368	- Difference between an array and a scalar, when src2 is constructed from Scalar or has the same
RyoheiHagimoto	0:0e0631af0305	369	number of elements as `src1.channels()`:
RyoheiHagimoto	0:0e0631af0305	370	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	371	- Difference between a scalar and an array, when src1 is constructed from Scalar or has the same
RyoheiHagimoto	0:0e0631af0305	372	number of elements as `src2.channels()`:
RyoheiHagimoto	0:0e0631af0305	373	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} - \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	374	- The reverse difference between a scalar and an array in the case of `SubRS`:
RyoheiHagimoto	0:0e0631af0305	375	\f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src2} - \texttt{src1}(I) ) \quad \texttt{if mask}(I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	376	where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each
RyoheiHagimoto	0:0e0631af0305	377	channel is processed independently.
RyoheiHagimoto	0:0e0631af0305	378
RyoheiHagimoto	0:0e0631af0305	379	The first function in the list above can be replaced with matrix expressions:
RyoheiHagimoto	0:0e0631af0305	380	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	381	dst = src1 - src2;
RyoheiHagimoto	0:0e0631af0305	382	dst -= src1; // equivalent to subtract(dst, src1, dst);
RyoheiHagimoto	0:0e0631af0305	383	@endcode
RyoheiHagimoto	0:0e0631af0305	384	The input arrays and the output array can all have the same or different depths. For example, you
RyoheiHagimoto	0:0e0631af0305	385	can subtract to 8-bit unsigned arrays and store the difference in a 16-bit signed array. Depth of
RyoheiHagimoto	0:0e0631af0305	386	the output array is determined by dtype parameter. In the second and third cases above, as well as
RyoheiHagimoto	0:0e0631af0305	387	in the first case, when src1.depth() == src2.depth(), dtype can be set to the default -1. In this
RyoheiHagimoto	0:0e0631af0305	388	case the output array will have the same depth as the input array, be it src1, src2 or both.
RyoheiHagimoto	0:0e0631af0305	389	@note Saturation is not applied when the output array has the depth CV_32S. You may even get
RyoheiHagimoto	0:0e0631af0305	390	result of an incorrect sign in the case of overflow.
RyoheiHagimoto	0:0e0631af0305	391	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	392	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	393	@param dst output array of the same size and the same number of channels as the input array.
RyoheiHagimoto	0:0e0631af0305	394	@param mask optional operation mask; this is an 8-bit single channel array that specifies elements
RyoheiHagimoto	0:0e0631af0305	395	of the output array to be changed.
RyoheiHagimoto	0:0e0631af0305	396	@param dtype optional depth of the output array
RyoheiHagimoto	0:0e0631af0305	397	@sa add, addWeighted, scaleAdd, Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	398	*/
RyoheiHagimoto	0:0e0631af0305	399	CV_EXPORTS_W void subtract(InputArray src1, InputArray src2, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	400	InputArray mask = noArray(), int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	401
RyoheiHagimoto	0:0e0631af0305	402
RyoheiHagimoto	0:0e0631af0305	403	/** @brief Calculates the per-element scaled product of two arrays.
RyoheiHagimoto	0:0e0631af0305	404
RyoheiHagimoto	0:0e0631af0305	405	The function multiply calculates the per-element product of two arrays:
RyoheiHagimoto	0:0e0631af0305	406
RyoheiHagimoto	0:0e0631af0305	407	\f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{scale} \cdot \texttt{src1} (I) \cdot \texttt{src2} (I))\f]
RyoheiHagimoto	0:0e0631af0305	408
RyoheiHagimoto	0:0e0631af0305	409	There is also a @ref MatrixExpressions -friendly variant of the first function. See Mat::mul .
RyoheiHagimoto	0:0e0631af0305	410
RyoheiHagimoto	0:0e0631af0305	411	For a not-per-element matrix product, see gemm .
RyoheiHagimoto	0:0e0631af0305	412
RyoheiHagimoto	0:0e0631af0305	413	@note Saturation is not applied when the output array has the depth
RyoheiHagimoto	0:0e0631af0305	414	CV_32S. You may even get result of an incorrect sign in the case of
RyoheiHagimoto	0:0e0631af0305	415	overflow.
RyoheiHagimoto	0:0e0631af0305	416	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	417	@param src2 second input array of the same size and the same type as src1.
RyoheiHagimoto	0:0e0631af0305	418	@param dst output array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	419	@param scale optional scale factor.
RyoheiHagimoto	0:0e0631af0305	420	@param dtype optional depth of the output array
RyoheiHagimoto	0:0e0631af0305	421	@sa add, subtract, divide, scaleAdd, addWeighted, accumulate, accumulateProduct, accumulateSquare,
RyoheiHagimoto	0:0e0631af0305	422	Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	423	*/
RyoheiHagimoto	0:0e0631af0305	424	CV_EXPORTS_W void multiply(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	425	OutputArray dst, double scale = 1, int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	426
RyoheiHagimoto	0:0e0631af0305	427	/** @brief Performs per-element division of two arrays or a scalar by an array.
RyoheiHagimoto	0:0e0631af0305	428
RyoheiHagimoto	0:0e0631af0305	429	The function cv::divide divides one array by another:
RyoheiHagimoto	0:0e0631af0305	430	\f[\texttt{dst(I) = saturate(src1(I)*scale/src2(I))}\f]
RyoheiHagimoto	0:0e0631af0305	431	or a scalar by an array when there is no src1 :
RyoheiHagimoto	0:0e0631af0305	432	\f[\texttt{dst(I) = saturate(scale/src2(I))}\f]
RyoheiHagimoto	0:0e0631af0305	433
RyoheiHagimoto	0:0e0631af0305	434	When src2(I) is zero, dst(I) will also be zero. Different channels of
RyoheiHagimoto	0:0e0631af0305	435	multi-channel arrays are processed independently.
RyoheiHagimoto	0:0e0631af0305	436
RyoheiHagimoto	0:0e0631af0305	437	@note Saturation is not applied when the output array has the depth CV_32S. You may even get
RyoheiHagimoto	0:0e0631af0305	438	result of an incorrect sign in the case of overflow.
RyoheiHagimoto	0:0e0631af0305	439	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	440	@param src2 second input array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	441	@param scale scalar factor.
RyoheiHagimoto	0:0e0631af0305	442	@param dst output array of the same size and type as src2.
RyoheiHagimoto	0:0e0631af0305	443	@param dtype optional depth of the output array; if -1, dst will have depth src2.depth(), but in
RyoheiHagimoto	0:0e0631af0305	444	case of an array-by-array division, you can only pass -1 when src1.depth()==src2.depth().
RyoheiHagimoto	0:0e0631af0305	445	@sa multiply, add, subtract
RyoheiHagimoto	0:0e0631af0305	446	*/
RyoheiHagimoto	0:0e0631af0305	447	CV_EXPORTS_W void divide(InputArray src1, InputArray src2, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	448	double scale = 1, int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	449
RyoheiHagimoto	0:0e0631af0305	450	/** @overload */
RyoheiHagimoto	0:0e0631af0305	451	CV_EXPORTS_W void divide(double scale, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	452	OutputArray dst, int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	453
RyoheiHagimoto	0:0e0631af0305	454	/** @brief Calculates the sum of a scaled array and another array.
RyoheiHagimoto	0:0e0631af0305	455
RyoheiHagimoto	0:0e0631af0305	456	The function scaleAdd is one of the classical primitive linear algebra operations, known as DAXPY
RyoheiHagimoto	0:0e0631af0305	457	or SAXPY in [BLAS](http://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms). It calculates
RyoheiHagimoto	0:0e0631af0305	458	the sum of a scaled array and another array:
RyoheiHagimoto	0:0e0631af0305	459	\f[\texttt{dst} (I)= \texttt{scale} \cdot \texttt{src1} (I) + \texttt{src2} (I)\f]
RyoheiHagimoto	0:0e0631af0305	460	The function can also be emulated with a matrix expression, for example:
RyoheiHagimoto	0:0e0631af0305	461	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	462	Mat A(3, 3, CV_64F);
RyoheiHagimoto	0:0e0631af0305	463	...
RyoheiHagimoto	0:0e0631af0305	464	A.row(0) = A.row(1)*2 + A.row(2);
RyoheiHagimoto	0:0e0631af0305	465	@endcode
RyoheiHagimoto	0:0e0631af0305	466	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	467	@param alpha scale factor for the first array.
RyoheiHagimoto	0:0e0631af0305	468	@param src2 second input array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	469	@param dst output array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	470	@sa add, addWeighted, subtract, Mat::dot, Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	471	*/
RyoheiHagimoto	0:0e0631af0305	472	CV_EXPORTS_W void scaleAdd(InputArray src1, double alpha, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	473
RyoheiHagimoto	0:0e0631af0305	474	/** @brief Calculates the weighted sum of two arrays.
RyoheiHagimoto	0:0e0631af0305	475
RyoheiHagimoto	0:0e0631af0305	476	The function addWeighted calculates the weighted sum of two arrays as follows:
RyoheiHagimoto	0:0e0631af0305	477	\f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{src1} (I)* \texttt{alpha} + \texttt{src2} (I)* \texttt{beta} + \texttt{gamma} )\f]
RyoheiHagimoto	0:0e0631af0305	478	where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each
RyoheiHagimoto	0:0e0631af0305	479	channel is processed independently.
RyoheiHagimoto	0:0e0631af0305	480	The function can be replaced with a matrix expression:
RyoheiHagimoto	0:0e0631af0305	481	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	482	dst = src1alpha + src2beta + gamma;
RyoheiHagimoto	0:0e0631af0305	483	@endcode
RyoheiHagimoto	0:0e0631af0305	484	@note Saturation is not applied when the output array has the depth CV_32S. You may even get
RyoheiHagimoto	0:0e0631af0305	485	result of an incorrect sign in the case of overflow.
RyoheiHagimoto	0:0e0631af0305	486	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	487	@param alpha weight of the first array elements.
RyoheiHagimoto	0:0e0631af0305	488	@param src2 second input array of the same size and channel number as src1.
RyoheiHagimoto	0:0e0631af0305	489	@param beta weight of the second array elements.
RyoheiHagimoto	0:0e0631af0305	490	@param gamma scalar added to each sum.
RyoheiHagimoto	0:0e0631af0305	491	@param dst output array that has the same size and number of channels as the input arrays.
RyoheiHagimoto	0:0e0631af0305	492	@param dtype optional depth of the output array; when both input arrays have the same depth, dtype
RyoheiHagimoto	0:0e0631af0305	493	can be set to -1, which will be equivalent to src1.depth().
RyoheiHagimoto	0:0e0631af0305	494	@sa add, subtract, scaleAdd, Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	495	*/
RyoheiHagimoto	0:0e0631af0305	496	CV_EXPORTS_W void addWeighted(InputArray src1, double alpha, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	497	double beta, double gamma, OutputArray dst, int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	498
RyoheiHagimoto	0:0e0631af0305	499	/** @brief Scales, calculates absolute values, and converts the result to 8-bit.
RyoheiHagimoto	0:0e0631af0305	500
RyoheiHagimoto	0:0e0631af0305	501	On each element of the input array, the function convertScaleAbs
RyoheiHagimoto	0:0e0631af0305	502	performs three operations sequentially: scaling, taking an absolute
RyoheiHagimoto	0:0e0631af0305	503	value, conversion to an unsigned 8-bit type:
RyoheiHagimoto	0:0e0631af0305	504	\f[\texttt{dst} (I)= \texttt{saturate\_cast<uchar>} (\| \texttt{src} (I)* \texttt{alpha} + \texttt{beta} \|)\f]
RyoheiHagimoto	0:0e0631af0305	505	In case of multi-channel arrays, the function processes each channel
RyoheiHagimoto	0:0e0631af0305	506	independently. When the output is not 8-bit, the operation can be
RyoheiHagimoto	0:0e0631af0305	507	emulated by calling the Mat::convertTo method (or by using matrix
RyoheiHagimoto	0:0e0631af0305	508	expressions) and then by calculating an absolute value of the result.
RyoheiHagimoto	0:0e0631af0305	509	For example:
RyoheiHagimoto	0:0e0631af0305	510	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	511	Mat_<float> A(30,30);
RyoheiHagimoto	0:0e0631af0305	512	randu(A, Scalar(-100), Scalar(100));
RyoheiHagimoto	0:0e0631af0305	513	Mat_<float> B = A*5 + 3;
RyoheiHagimoto	0:0e0631af0305	514	B = abs(B);
RyoheiHagimoto	0:0e0631af0305	515	// Mat_<float> B = abs(A*5+3) will also do the job,
RyoheiHagimoto	0:0e0631af0305	516	// but it will allocate a temporary matrix
RyoheiHagimoto	0:0e0631af0305	517	@endcode
RyoheiHagimoto	0:0e0631af0305	518	@param src input array.
RyoheiHagimoto	0:0e0631af0305	519	@param dst output array.
RyoheiHagimoto	0:0e0631af0305	520	@param alpha optional scale factor.
RyoheiHagimoto	0:0e0631af0305	521	@param beta optional delta added to the scaled values.
RyoheiHagimoto	0:0e0631af0305	522	@sa Mat::convertTo, cv::abs(const Mat&)
RyoheiHagimoto	0:0e0631af0305	523	*/
RyoheiHagimoto	0:0e0631af0305	524	CV_EXPORTS_W void convertScaleAbs(InputArray src, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	525	double alpha = 1, double beta = 0);
RyoheiHagimoto	0:0e0631af0305	526
RyoheiHagimoto	0:0e0631af0305	527	/** @brief Converts an array to half precision floating number.
RyoheiHagimoto	0:0e0631af0305	528
RyoheiHagimoto	0:0e0631af0305	529	This function converts FP32 (single precision floating point) from/to FP16 (half precision floating point). The input array has to have type of CV_32F or
RyoheiHagimoto	0:0e0631af0305	530	CV_16S to represent the bit depth. If the input array is neither of them, the function will raise an error.
RyoheiHagimoto	0:0e0631af0305	531	The format of half precision floating point is defined in IEEE 754-2008.
RyoheiHagimoto	0:0e0631af0305	532
RyoheiHagimoto	0:0e0631af0305	533	@param src input array.
RyoheiHagimoto	0:0e0631af0305	534	@param dst output array.
RyoheiHagimoto	0:0e0631af0305	535	*/
RyoheiHagimoto	0:0e0631af0305	536	CV_EXPORTS_W void convertFp16(InputArray src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	537
RyoheiHagimoto	0:0e0631af0305	538	/** @brief Performs a look-up table transform of an array.
RyoheiHagimoto	0:0e0631af0305	539
RyoheiHagimoto	0:0e0631af0305	540	The function LUT fills the output array with values from the look-up table. Indices of the entries
RyoheiHagimoto	0:0e0631af0305	541	are taken from the input array. That is, the function processes each element of src as follows:
RyoheiHagimoto	0:0e0631af0305	542	\f[\texttt{dst} (I) \leftarrow \texttt{lut(src(I) + d)}\f]
RyoheiHagimoto	0:0e0631af0305	543	where
RyoheiHagimoto	0:0e0631af0305	544	\f[d = \fork{0}{if $\texttt{src}$ has depth $\texttt{CV_8U}$}{128}{if $\texttt{src}$ has depth $\texttt{CV_8S}$}\f]
RyoheiHagimoto	0:0e0631af0305	545	@param src input array of 8-bit elements.
RyoheiHagimoto	0:0e0631af0305	546	@param lut look-up table of 256 elements; in case of multi-channel input array, the table should
RyoheiHagimoto	0:0e0631af0305	547	either have a single channel (in this case the same table is used for all channels) or the same
RyoheiHagimoto	0:0e0631af0305	548	number of channels as in the input array.
RyoheiHagimoto	0:0e0631af0305	549	@param dst output array of the same size and number of channels as src, and the same depth as lut.
RyoheiHagimoto	0:0e0631af0305	550	@sa convertScaleAbs, Mat::convertTo
RyoheiHagimoto	0:0e0631af0305	551	*/
RyoheiHagimoto	0:0e0631af0305	552	CV_EXPORTS_W void LUT(InputArray src, InputArray lut, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	553
RyoheiHagimoto	0:0e0631af0305	554	/** @brief Calculates the sum of array elements.
RyoheiHagimoto	0:0e0631af0305	555
RyoheiHagimoto	0:0e0631af0305	556	The function cv::sum calculates and returns the sum of array elements,
RyoheiHagimoto	0:0e0631af0305	557	independently for each channel.
RyoheiHagimoto	0:0e0631af0305	558	@param src input array that must have from 1 to 4 channels.
RyoheiHagimoto	0:0e0631af0305	559	@sa countNonZero, mean, meanStdDev, norm, minMaxLoc, reduce
RyoheiHagimoto	0:0e0631af0305	560	*/
RyoheiHagimoto	0:0e0631af0305	561	CV_EXPORTS_AS(sumElems) Scalar sum(InputArray src);
RyoheiHagimoto	0:0e0631af0305	562
RyoheiHagimoto	0:0e0631af0305	563	/** @brief Counts non-zero array elements.
RyoheiHagimoto	0:0e0631af0305	564
RyoheiHagimoto	0:0e0631af0305	565	The function returns the number of non-zero elements in src :
RyoheiHagimoto	0:0e0631af0305	566	\f[\sum _{I: \; \texttt{src} (I) \ne0 } 1\f]
RyoheiHagimoto	0:0e0631af0305	567	@param src single-channel array.
RyoheiHagimoto	0:0e0631af0305	568	@sa mean, meanStdDev, norm, minMaxLoc, calcCovarMatrix
RyoheiHagimoto	0:0e0631af0305	569	*/
RyoheiHagimoto	0:0e0631af0305	570	CV_EXPORTS_W int countNonZero( InputArray src );
RyoheiHagimoto	0:0e0631af0305	571
RyoheiHagimoto	0:0e0631af0305	572	/** @brief Returns the list of locations of non-zero pixels
RyoheiHagimoto	0:0e0631af0305	573
RyoheiHagimoto	0:0e0631af0305	574	Given a binary matrix (likely returned from an operation such
RyoheiHagimoto	0:0e0631af0305	575	as threshold(), compare(), >, ==, etc, return all of
RyoheiHagimoto	0:0e0631af0305	576	the non-zero indices as a cv::Mat or std::vector<cv::Point> (x,y)
RyoheiHagimoto	0:0e0631af0305	577	For example:
RyoheiHagimoto	0:0e0631af0305	578	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	579	cv::Mat binaryImage; // input, binary image
RyoheiHagimoto	0:0e0631af0305	580	cv::Mat locations; // output, locations of non-zero pixels
RyoheiHagimoto	0:0e0631af0305	581	cv::findNonZero(binaryImage, locations);
RyoheiHagimoto	0:0e0631af0305	582
RyoheiHagimoto	0:0e0631af0305	583	// access pixel coordinates
RyoheiHagimoto	0:0e0631af0305	584	Point pnt = locations.at<Point>(i);
RyoheiHagimoto	0:0e0631af0305	585	@endcode
RyoheiHagimoto	0:0e0631af0305	586	or
RyoheiHagimoto	0:0e0631af0305	587	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	588	cv::Mat binaryImage; // input, binary image
RyoheiHagimoto	0:0e0631af0305	589	vector<Point> locations; // output, locations of non-zero pixels
RyoheiHagimoto	0:0e0631af0305	590	cv::findNonZero(binaryImage, locations);
RyoheiHagimoto	0:0e0631af0305	591
RyoheiHagimoto	0:0e0631af0305	592	// access pixel coordinates
RyoheiHagimoto	0:0e0631af0305	593	Point pnt = locations[i];
RyoheiHagimoto	0:0e0631af0305	594	@endcode
RyoheiHagimoto	0:0e0631af0305	595	@param src single-channel array (type CV_8UC1)
RyoheiHagimoto	0:0e0631af0305	596	@param idx the output array, type of cv::Mat or std::vector<Point>, corresponding to non-zero indices in the input
RyoheiHagimoto	0:0e0631af0305	597	*/
RyoheiHagimoto	0:0e0631af0305	598	CV_EXPORTS_W void findNonZero( InputArray src, OutputArray idx );
RyoheiHagimoto	0:0e0631af0305	599
RyoheiHagimoto	0:0e0631af0305	600	/** @brief Calculates an average (mean) of array elements.
RyoheiHagimoto	0:0e0631af0305	601
RyoheiHagimoto	0:0e0631af0305	602	The function cv::mean calculates the mean value M of array elements,
RyoheiHagimoto	0:0e0631af0305	603	independently for each channel, and return it:
RyoheiHagimoto	0:0e0631af0305	604	\f[\begin{array}{l} N = \sum _{I: \; \texttt{mask} (I) \ne 0} 1 \\ M_c = \left ( \sum _{I: \; \texttt{mask} (I) \ne 0}{ \texttt{mtx} (I)_c} \right )/N \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	605	When all the mask elements are 0's, the function returns Scalar::all(0)
RyoheiHagimoto	0:0e0631af0305	606	@param src input array that should have from 1 to 4 channels so that the result can be stored in
RyoheiHagimoto	0:0e0631af0305	607	Scalar_ .
RyoheiHagimoto	0:0e0631af0305	608	@param mask optional operation mask.
RyoheiHagimoto	0:0e0631af0305	609	@sa countNonZero, meanStdDev, norm, minMaxLoc
RyoheiHagimoto	0:0e0631af0305	610	*/
RyoheiHagimoto	0:0e0631af0305	611	CV_EXPORTS_W Scalar mean(InputArray src, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	612
RyoheiHagimoto	0:0e0631af0305	613	/** Calculates a mean and standard deviation of array elements.
RyoheiHagimoto	0:0e0631af0305	614
RyoheiHagimoto	0:0e0631af0305	615	The function cv::meanStdDev calculates the mean and the standard deviation M
RyoheiHagimoto	0:0e0631af0305	616	of array elements independently for each channel and returns it via the
RyoheiHagimoto	0:0e0631af0305	617	output parameters:
RyoheiHagimoto	0:0e0631af0305	618	\f[\begin{array}{l} N = \sum _{I, \texttt{mask} (I) \ne 0} 1 \\ \texttt{mean} _c = \frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \texttt{src} (I)_c}{N} \\ \texttt{stddev} _c = \sqrt{\frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \left ( \texttt{src} (I)_c - \texttt{mean} _c \right )^2}{N}} \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	619	When all the mask elements are 0's, the function returns
RyoheiHagimoto	0:0e0631af0305	620	mean=stddev=Scalar::all(0).
RyoheiHagimoto	0:0e0631af0305	621	@note The calculated standard deviation is only the diagonal of the
RyoheiHagimoto	0:0e0631af0305	622	complete normalized covariance matrix. If the full matrix is needed, you
RyoheiHagimoto	0:0e0631af0305	623	can reshape the multi-channel array M x N to the single-channel array
RyoheiHagimoto	0:0e0631af0305	624	M\*N x mtx.channels() (only possible when the matrix is continuous) and
RyoheiHagimoto	0:0e0631af0305	625	then pass the matrix to calcCovarMatrix .
RyoheiHagimoto	0:0e0631af0305	626	@param src input array that should have from 1 to 4 channels so that the results can be stored in
RyoheiHagimoto	0:0e0631af0305	627	Scalar_ 's.
RyoheiHagimoto	0:0e0631af0305	628	@param mean output parameter: calculated mean value.
RyoheiHagimoto	0:0e0631af0305	629	@param stddev output parameter: calculateded standard deviation.
RyoheiHagimoto	0:0e0631af0305	630	@param mask optional operation mask.
RyoheiHagimoto	0:0e0631af0305	631	@sa countNonZero, mean, norm, minMaxLoc, calcCovarMatrix
RyoheiHagimoto	0:0e0631af0305	632	*/
RyoheiHagimoto	0:0e0631af0305	633	CV_EXPORTS_W void meanStdDev(InputArray src, OutputArray mean, OutputArray stddev,
RyoheiHagimoto	0:0e0631af0305	634	InputArray mask=noArray());
RyoheiHagimoto	0:0e0631af0305	635
RyoheiHagimoto	0:0e0631af0305	636	/** @brief Calculates an absolute array norm, an absolute difference norm, or a
RyoheiHagimoto	0:0e0631af0305	637	relative difference norm.
RyoheiHagimoto	0:0e0631af0305	638
RyoheiHagimoto	0:0e0631af0305	639	The function cv::norm calculates an absolute norm of src1 (when there is no
RyoheiHagimoto	0:0e0631af0305	640	src2 ):
RyoheiHagimoto	0:0e0631af0305	641
RyoheiHagimoto	0:0e0631af0305	642	\f[norm = \forkthree{\\|\texttt{src1}\\|_{L_{\infty}} = \max _I \| \texttt{src1} (I)\|}{if $\texttt{normType} = \texttt{NORM_INF}$ }
RyoheiHagimoto	0:0e0631af0305	643	{ \\| \texttt{src1} \\| _{L_1} = \sum _I \| \texttt{src1} (I)\|}{if $\texttt{normType} = \texttt{NORM_L1}$ }
RyoheiHagimoto	0:0e0631af0305	644	{ \\| \texttt{src1} \\| _{L_2} = \sqrt{\sum_I \texttt{src1}(I)^2} }{if $\texttt{normType} = \texttt{NORM_L2}$ }\f]
RyoheiHagimoto	0:0e0631af0305	645
RyoheiHagimoto	0:0e0631af0305	646	or an absolute or relative difference norm if src2 is there:
RyoheiHagimoto	0:0e0631af0305	647
RyoheiHagimoto	0:0e0631af0305	648	\f[norm = \forkthree{\\|\texttt{src1}-\texttt{src2}\\|_{L_{\infty}} = \max _I \| \texttt{src1} (I) - \texttt{src2} (I)\|}{if $\texttt{normType} = \texttt{NORM_INF}$ }
RyoheiHagimoto	0:0e0631af0305	649	{ \\| \texttt{src1} - \texttt{src2} \\| _{L_1} = \sum _I \| \texttt{src1} (I) - \texttt{src2} (I)\|}{if $\texttt{normType} = \texttt{NORM_L1}$ }
RyoheiHagimoto	0:0e0631af0305	650	{ \\| \texttt{src1} - \texttt{src2} \\| _{L_2} = \sqrt{\sum_I (\texttt{src1}(I) - \texttt{src2}(I))^2} }{if $\texttt{normType} = \texttt{NORM_L2}$ }\f]
RyoheiHagimoto	0:0e0631af0305	651
RyoheiHagimoto	0:0e0631af0305	652	or
RyoheiHagimoto	0:0e0631af0305	653
RyoheiHagimoto	0:0e0631af0305	654	\f[norm = \forkthree{\frac{\\|\texttt{src1}-\texttt{src2}\\|_{L_{\infty}} }{\\|\texttt{src2}\\|_{L_{\infty}} }}{if $\texttt{normType} = \texttt{NORM_RELATIVE_INF}$ }
RyoheiHagimoto	0:0e0631af0305	655	{ \frac{\\|\texttt{src1}-\texttt{src2}\\|_{L_1} }{\\|\texttt{src2}\\|_{L_1}} }{if $\texttt{normType} = \texttt{NORM_RELATIVE_L1}$ }
RyoheiHagimoto	0:0e0631af0305	656	{ \frac{\\|\texttt{src1}-\texttt{src2}\\|_{L_2} }{\\|\texttt{src2}\\|_{L_2}} }{if $\texttt{normType} = \texttt{NORM_RELATIVE_L2}$ }\f]
RyoheiHagimoto	0:0e0631af0305	657
RyoheiHagimoto	0:0e0631af0305	658	The function cv::norm returns the calculated norm.
RyoheiHagimoto	0:0e0631af0305	659
RyoheiHagimoto	0:0e0631af0305	660	When the mask parameter is specified and it is not empty, the norm is
RyoheiHagimoto	0:0e0631af0305	661	calculated only over the region specified by the mask.
RyoheiHagimoto	0:0e0631af0305	662
RyoheiHagimoto	0:0e0631af0305	663	A multi-channel input arrays are treated as a single-channel, that is,
RyoheiHagimoto	0:0e0631af0305	664	the results for all channels are combined.
RyoheiHagimoto	0:0e0631af0305	665
RyoheiHagimoto	0:0e0631af0305	666	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	667	@param normType type of the norm (see cv::NormTypes).
RyoheiHagimoto	0:0e0631af0305	668	@param mask optional operation mask; it must have the same size as src1 and CV_8UC1 type.
RyoheiHagimoto	0:0e0631af0305	669	*/
RyoheiHagimoto	0:0e0631af0305	670	CV_EXPORTS_W double norm(InputArray src1, int normType = NORM_L2, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	671
RyoheiHagimoto	0:0e0631af0305	672	/** @overload
RyoheiHagimoto	0:0e0631af0305	673	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	674	@param src2 second input array of the same size and the same type as src1.
RyoheiHagimoto	0:0e0631af0305	675	@param normType type of the norm (cv::NormTypes).
RyoheiHagimoto	0:0e0631af0305	676	@param mask optional operation mask; it must have the same size as src1 and CV_8UC1 type.
RyoheiHagimoto	0:0e0631af0305	677	*/
RyoheiHagimoto	0:0e0631af0305	678	CV_EXPORTS_W double norm(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	679	int normType = NORM_L2, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	680	/** @overload
RyoheiHagimoto	0:0e0631af0305	681	@param src first input array.
RyoheiHagimoto	0:0e0631af0305	682	@param normType type of the norm (see cv::NormTypes).
RyoheiHagimoto	0:0e0631af0305	683	*/
RyoheiHagimoto	0:0e0631af0305	684	CV_EXPORTS double norm( const SparseMat& src, int normType );
RyoheiHagimoto	0:0e0631af0305	685
RyoheiHagimoto	0:0e0631af0305	686	/** @brief computes PSNR image/video quality metric
RyoheiHagimoto	0:0e0631af0305	687
RyoheiHagimoto	0:0e0631af0305	688	see http://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio for details
RyoheiHagimoto	0:0e0631af0305	689	@todo document
RyoheiHagimoto	0:0e0631af0305	690	*/
RyoheiHagimoto	0:0e0631af0305	691	CV_EXPORTS_W double PSNR(InputArray src1, InputArray src2);
RyoheiHagimoto	0:0e0631af0305	692
RyoheiHagimoto	0:0e0631af0305	693	/** @brief naive nearest neighbor finder
RyoheiHagimoto	0:0e0631af0305	694
RyoheiHagimoto	0:0e0631af0305	695	see http://en.wikipedia.org/wiki/Nearest_neighbor_search
RyoheiHagimoto	0:0e0631af0305	696	@todo document
RyoheiHagimoto	0:0e0631af0305	697	*/
RyoheiHagimoto	0:0e0631af0305	698	CV_EXPORTS_W void batchDistance(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	699	OutputArray dist, int dtype, OutputArray nidx,
RyoheiHagimoto	0:0e0631af0305	700	int normType = NORM_L2, int K = 0,
RyoheiHagimoto	0:0e0631af0305	701	InputArray mask = noArray(), int update = 0,
RyoheiHagimoto	0:0e0631af0305	702	bool crosscheck = false);
RyoheiHagimoto	0:0e0631af0305	703
RyoheiHagimoto	0:0e0631af0305	704	/** @brief Normalizes the norm or value range of an array.
RyoheiHagimoto	0:0e0631af0305	705
RyoheiHagimoto	0:0e0631af0305	706	The function cv::normalize normalizes scale and shift the input array elements so that
RyoheiHagimoto	0:0e0631af0305	707	\f[\\| \texttt{dst} \\| _{L_p}= \texttt{alpha}\f]
RyoheiHagimoto	0:0e0631af0305	708	(where p=Inf, 1 or 2) when normType=NORM_INF, NORM_L1, or NORM_L2, respectively; or so that
RyoheiHagimoto	0:0e0631af0305	709	\f[\min _I \texttt{dst} (I)= \texttt{alpha} , \, \, \max _I \texttt{dst} (I)= \texttt{beta}\f]
RyoheiHagimoto	0:0e0631af0305	710
RyoheiHagimoto	0:0e0631af0305	711	when normType=NORM_MINMAX (for dense arrays only). The optional mask specifies a sub-array to be
RyoheiHagimoto	0:0e0631af0305	712	normalized. This means that the norm or min-n-max are calculated over the sub-array, and then this
RyoheiHagimoto	0:0e0631af0305	713	sub-array is modified to be normalized. If you want to only use the mask to calculate the norm or
RyoheiHagimoto	0:0e0631af0305	714	min-max but modify the whole array, you can use norm and Mat::convertTo.
RyoheiHagimoto	0:0e0631af0305	715
RyoheiHagimoto	0:0e0631af0305	716	In case of sparse matrices, only the non-zero values are analyzed and transformed. Because of this,
RyoheiHagimoto	0:0e0631af0305	717	the range transformation for sparse matrices is not allowed since it can shift the zero level.
RyoheiHagimoto	0:0e0631af0305	718
RyoheiHagimoto	0:0e0631af0305	719	Possible usage with some positive example data:
RyoheiHagimoto	0:0e0631af0305	720	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	721	vector<double> positiveData = { 2.0, 8.0, 10.0 };
RyoheiHagimoto	0:0e0631af0305	722	vector<double> normalizedData_l1, normalizedData_l2, normalizedData_inf, normalizedData_minmax;
RyoheiHagimoto	0:0e0631af0305	723
RyoheiHagimoto	0:0e0631af0305	724	// Norm to probability (total count)
RyoheiHagimoto	0:0e0631af0305	725	// sum(numbers) = 20.0
RyoheiHagimoto	0:0e0631af0305	726	// 2.0 0.1 (2.0/20.0)
RyoheiHagimoto	0:0e0631af0305	727	// 8.0 0.4 (8.0/20.0)
RyoheiHagimoto	0:0e0631af0305	728	// 10.0 0.5 (10.0/20.0)
RyoheiHagimoto	0:0e0631af0305	729	normalize(positiveData, normalizedData_l1, 1.0, 0.0, NORM_L1);
RyoheiHagimoto	0:0e0631af0305	730
RyoheiHagimoto	0:0e0631af0305	731	// Norm to unit vector: \|\|positiveData\|\| = 1.0
RyoheiHagimoto	0:0e0631af0305	732	// 2.0 0.15
RyoheiHagimoto	0:0e0631af0305	733	// 8.0 0.62
RyoheiHagimoto	0:0e0631af0305	734	// 10.0 0.77
RyoheiHagimoto	0:0e0631af0305	735	normalize(positiveData, normalizedData_l2, 1.0, 0.0, NORM_L2);
RyoheiHagimoto	0:0e0631af0305	736
RyoheiHagimoto	0:0e0631af0305	737	// Norm to max element
RyoheiHagimoto	0:0e0631af0305	738	// 2.0 0.2 (2.0/10.0)
RyoheiHagimoto	0:0e0631af0305	739	// 8.0 0.8 (8.0/10.0)
RyoheiHagimoto	0:0e0631af0305	740	// 10.0 1.0 (10.0/10.0)
RyoheiHagimoto	0:0e0631af0305	741	normalize(positiveData, normalizedData_inf, 1.0, 0.0, NORM_INF);
RyoheiHagimoto	0:0e0631af0305	742
RyoheiHagimoto	0:0e0631af0305	743	// Norm to range [0.0;1.0]
RyoheiHagimoto	0:0e0631af0305	744	// 2.0 0.0 (shift to left border)
RyoheiHagimoto	0:0e0631af0305	745	// 8.0 0.75 (6.0/8.0)
RyoheiHagimoto	0:0e0631af0305	746	// 10.0 1.0 (shift to right border)
RyoheiHagimoto	0:0e0631af0305	747	normalize(positiveData, normalizedData_minmax, 1.0, 0.0, NORM_MINMAX);
RyoheiHagimoto	0:0e0631af0305	748	@endcode
RyoheiHagimoto	0:0e0631af0305	749
RyoheiHagimoto	0:0e0631af0305	750	@param src input array.
RyoheiHagimoto	0:0e0631af0305	751	@param dst output array of the same size as src .
RyoheiHagimoto	0:0e0631af0305	752	@param alpha norm value to normalize to or the lower range boundary in case of the range
RyoheiHagimoto	0:0e0631af0305	753	normalization.
RyoheiHagimoto	0:0e0631af0305	754	@param beta upper range boundary in case of the range normalization; it is not used for the norm
RyoheiHagimoto	0:0e0631af0305	755	normalization.
RyoheiHagimoto	0:0e0631af0305	756	@param norm_type normalization type (see cv::NormTypes).
RyoheiHagimoto	0:0e0631af0305	757	@param dtype when negative, the output array has the same type as src; otherwise, it has the same
RyoheiHagimoto	0:0e0631af0305	758	number of channels as src and the depth =CV_MAT_DEPTH(dtype).
RyoheiHagimoto	0:0e0631af0305	759	@param mask optional operation mask.
RyoheiHagimoto	0:0e0631af0305	760	@sa norm, Mat::convertTo, SparseMat::convertTo
RyoheiHagimoto	0:0e0631af0305	761	*/
RyoheiHagimoto	0:0e0631af0305	762	CV_EXPORTS_W void normalize( InputArray src, InputOutputArray dst, double alpha = 1, double beta = 0,
RyoheiHagimoto	0:0e0631af0305	763	int norm_type = NORM_L2, int dtype = -1, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	764
RyoheiHagimoto	0:0e0631af0305	765	/** @overload
RyoheiHagimoto	0:0e0631af0305	766	@param src input array.
RyoheiHagimoto	0:0e0631af0305	767	@param dst output array of the same size as src .
RyoheiHagimoto	0:0e0631af0305	768	@param alpha norm value to normalize to or the lower range boundary in case of the range
RyoheiHagimoto	0:0e0631af0305	769	normalization.
RyoheiHagimoto	0:0e0631af0305	770	@param normType normalization type (see cv::NormTypes).
RyoheiHagimoto	0:0e0631af0305	771	*/
RyoheiHagimoto	0:0e0631af0305	772	CV_EXPORTS void normalize( const SparseMat& src, SparseMat& dst, double alpha, int normType );
RyoheiHagimoto	0:0e0631af0305	773
RyoheiHagimoto	0:0e0631af0305	774	/** @brief Finds the global minimum and maximum in an array.
RyoheiHagimoto	0:0e0631af0305	775
RyoheiHagimoto	0:0e0631af0305	776	The function cv::minMaxLoc finds the minimum and maximum element values and their positions. The
RyoheiHagimoto	0:0e0631af0305	777	extremums are searched across the whole array or, if mask is not an empty array, in the specified
RyoheiHagimoto	0:0e0631af0305	778	array region.
RyoheiHagimoto	0:0e0631af0305	779
RyoheiHagimoto	0:0e0631af0305	780	The function do not work with multi-channel arrays. If you need to find minimum or maximum
RyoheiHagimoto	0:0e0631af0305	781	elements across all the channels, use Mat::reshape first to reinterpret the array as
RyoheiHagimoto	0:0e0631af0305	782	single-channel. Or you may extract the particular channel using either extractImageCOI , or
RyoheiHagimoto	0:0e0631af0305	783	mixChannels , or split .
RyoheiHagimoto	0:0e0631af0305	784	@param src input single-channel array.
RyoheiHagimoto	0:0e0631af0305	785	@param minVal pointer to the returned minimum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	786	@param maxVal pointer to the returned maximum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	787	@param minLoc pointer to the returned minimum location (in 2D case); NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	788	@param maxLoc pointer to the returned maximum location (in 2D case); NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	789	@param mask optional mask used to select a sub-array.
RyoheiHagimoto	0:0e0631af0305	790	@sa max, min, compare, inRange, extractImageCOI, mixChannels, split, Mat::reshape
RyoheiHagimoto	0:0e0631af0305	791	*/
RyoheiHagimoto	0:0e0631af0305	792	CV_EXPORTS_W void minMaxLoc(InputArray src, CV_OUT double* minVal,
RyoheiHagimoto	0:0e0631af0305	793	CV_OUT double* maxVal = 0, CV_OUT Point* minLoc = 0,
RyoheiHagimoto	0:0e0631af0305	794	CV_OUT Point* maxLoc = 0, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	795
RyoheiHagimoto	0:0e0631af0305	796
RyoheiHagimoto	0:0e0631af0305	797	/** @brief Finds the global minimum and maximum in an array
RyoheiHagimoto	0:0e0631af0305	798
RyoheiHagimoto	0:0e0631af0305	799	The function cv::minMaxIdx finds the minimum and maximum element values and their positions. The
RyoheiHagimoto	0:0e0631af0305	800	extremums are searched across the whole array or, if mask is not an empty array, in the specified
RyoheiHagimoto	0:0e0631af0305	801	array region. The function does not work with multi-channel arrays. If you need to find minimum or
RyoheiHagimoto	0:0e0631af0305	802	maximum elements across all the channels, use Mat::reshape first to reinterpret the array as
RyoheiHagimoto	0:0e0631af0305	803	single-channel. Or you may extract the particular channel using either extractImageCOI , or
RyoheiHagimoto	0:0e0631af0305	804	mixChannels , or split . In case of a sparse matrix, the minimum is found among non-zero elements
RyoheiHagimoto	0:0e0631af0305	805	only.
RyoheiHagimoto	0:0e0631af0305	806	@note When minIdx is not NULL, it must have at least 2 elements (as well as maxIdx), even if src is
RyoheiHagimoto	0:0e0631af0305	807	a single-row or single-column matrix. In OpenCV (following MATLAB) each array has at least 2
RyoheiHagimoto	0:0e0631af0305	808	dimensions, i.e. single-column matrix is Mx1 matrix (and therefore minIdx/maxIdx will be
RyoheiHagimoto	0:0e0631af0305	809	(i1,0)/(i2,0)) and single-row matrix is 1xN matrix (and therefore minIdx/maxIdx will be
RyoheiHagimoto	0:0e0631af0305	810	(0,j1)/(0,j2)).
RyoheiHagimoto	0:0e0631af0305	811	@param src input single-channel array.
RyoheiHagimoto	0:0e0631af0305	812	@param minVal pointer to the returned minimum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	813	@param maxVal pointer to the returned maximum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	814	@param minIdx pointer to the returned minimum location (in nD case); NULL is used if not required;
RyoheiHagimoto	0:0e0631af0305	815	Otherwise, it must point to an array of src.dims elements, the coordinates of the minimum element
RyoheiHagimoto	0:0e0631af0305	816	in each dimension are stored there sequentially.
RyoheiHagimoto	0:0e0631af0305	817	@param maxIdx pointer to the returned maximum location (in nD case). NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	818	@param mask specified array region
RyoheiHagimoto	0:0e0631af0305	819	*/
RyoheiHagimoto	0:0e0631af0305	820	CV_EXPORTS void minMaxIdx(InputArray src, double* minVal, double* maxVal = 0,
RyoheiHagimoto	0:0e0631af0305	821	int* minIdx = 0, int* maxIdx = 0, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	822
RyoheiHagimoto	0:0e0631af0305	823	/** @overload
RyoheiHagimoto	0:0e0631af0305	824	@param a input single-channel array.
RyoheiHagimoto	0:0e0631af0305	825	@param minVal pointer to the returned minimum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	826	@param maxVal pointer to the returned maximum value; NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	827	@param minIdx pointer to the returned minimum location (in nD case); NULL is used if not required;
RyoheiHagimoto	0:0e0631af0305	828	Otherwise, it must point to an array of src.dims elements, the coordinates of the minimum element
RyoheiHagimoto	0:0e0631af0305	829	in each dimension are stored there sequentially.
RyoheiHagimoto	0:0e0631af0305	830	@param maxIdx pointer to the returned maximum location (in nD case). NULL is used if not required.
RyoheiHagimoto	0:0e0631af0305	831	*/
RyoheiHagimoto	0:0e0631af0305	832	CV_EXPORTS void minMaxLoc(const SparseMat& a, double* minVal,
RyoheiHagimoto	0:0e0631af0305	833	double* maxVal, int* minIdx = 0, int* maxIdx = 0);
RyoheiHagimoto	0:0e0631af0305	834
RyoheiHagimoto	0:0e0631af0305	835	/** @brief Reduces a matrix to a vector.
RyoheiHagimoto	0:0e0631af0305	836
RyoheiHagimoto	0:0e0631af0305	837	The function cv::reduce reduces the matrix to a vector by treating the matrix rows/columns as a set of
RyoheiHagimoto	0:0e0631af0305	838	1D vectors and performing the specified operation on the vectors until a single row/column is
RyoheiHagimoto	0:0e0631af0305	839	obtained. For example, the function can be used to compute horizontal and vertical projections of a
RyoheiHagimoto	0:0e0631af0305	840	raster image. In case of REDUCE_MAX and REDUCE_MIN , the output image should have the same type as the source one.
RyoheiHagimoto	0:0e0631af0305	841	In case of REDUCE_SUM and REDUCE_AVG , the output may have a larger element bit-depth to preserve accuracy.
RyoheiHagimoto	0:0e0631af0305	842	And multi-channel arrays are also supported in these two reduction modes.
RyoheiHagimoto	0:0e0631af0305	843	@param src input 2D matrix.
RyoheiHagimoto	0:0e0631af0305	844	@param dst output vector. Its size and type is defined by dim and dtype parameters.
RyoheiHagimoto	0:0e0631af0305	845	@param dim dimension index along which the matrix is reduced. 0 means that the matrix is reduced to
RyoheiHagimoto	0:0e0631af0305	846	a single row. 1 means that the matrix is reduced to a single column.
RyoheiHagimoto	0:0e0631af0305	847	@param rtype reduction operation that could be one of cv::ReduceTypes
RyoheiHagimoto	0:0e0631af0305	848	@param dtype when negative, the output vector will have the same type as the input matrix,
RyoheiHagimoto	0:0e0631af0305	849	otherwise, its type will be CV_MAKE_TYPE(CV_MAT_DEPTH(dtype), src.channels()).
RyoheiHagimoto	0:0e0631af0305	850	@sa repeat
RyoheiHagimoto	0:0e0631af0305	851	*/
RyoheiHagimoto	0:0e0631af0305	852	CV_EXPORTS_W void reduce(InputArray src, OutputArray dst, int dim, int rtype, int dtype = -1);
RyoheiHagimoto	0:0e0631af0305	853
RyoheiHagimoto	0:0e0631af0305	854	/** @brief Creates one multi-channel array out of several single-channel ones.
RyoheiHagimoto	0:0e0631af0305	855
RyoheiHagimoto	0:0e0631af0305	856	The function cv::merge merges several arrays to make a single multi-channel array. That is, each
RyoheiHagimoto	0:0e0631af0305	857	element of the output array will be a concatenation of the elements of the input arrays, where
RyoheiHagimoto	0:0e0631af0305	858	elements of i-th input array are treated as mv[i].channels()-element vectors.
RyoheiHagimoto	0:0e0631af0305	859
RyoheiHagimoto	0:0e0631af0305	860	The function cv::split does the reverse operation. If you need to shuffle channels in some other
RyoheiHagimoto	0:0e0631af0305	861	advanced way, use cv::mixChannels.
RyoheiHagimoto	0:0e0631af0305	862	@param mv input array of matrices to be merged; all the matrices in mv must have the same
RyoheiHagimoto	0:0e0631af0305	863	size and the same depth.
RyoheiHagimoto	0:0e0631af0305	864	@param count number of input matrices when mv is a plain C array; it must be greater than zero.
RyoheiHagimoto	0:0e0631af0305	865	@param dst output array of the same size and the same depth as mv[0]; The number of channels will
RyoheiHagimoto	0:0e0631af0305	866	be equal to the parameter count.
RyoheiHagimoto	0:0e0631af0305	867	@sa mixChannels, split, Mat::reshape
RyoheiHagimoto	0:0e0631af0305	868	*/
RyoheiHagimoto	0:0e0631af0305	869	CV_EXPORTS void merge(const Mat* mv, size_t count, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	870
RyoheiHagimoto	0:0e0631af0305	871	/** @overload
RyoheiHagimoto	0:0e0631af0305	872	@param mv input vector of matrices to be merged; all the matrices in mv must have the same
RyoheiHagimoto	0:0e0631af0305	873	size and the same depth.
RyoheiHagimoto	0:0e0631af0305	874	@param dst output array of the same size and the same depth as mv[0]; The number of channels will
RyoheiHagimoto	0:0e0631af0305	875	be the total number of channels in the matrix array.
RyoheiHagimoto	0:0e0631af0305	876	*/
RyoheiHagimoto	0:0e0631af0305	877	CV_EXPORTS_W void merge(InputArrayOfArrays mv, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	878
RyoheiHagimoto	0:0e0631af0305	879	/** @brief Divides a multi-channel array into several single-channel arrays.
RyoheiHagimoto	0:0e0631af0305	880
RyoheiHagimoto	0:0e0631af0305	881	The function cv::split splits a multi-channel array into separate single-channel arrays:
RyoheiHagimoto	0:0e0631af0305	882	\f[\texttt{mv} [c](I) = \texttt{src} (I)_c\f]
RyoheiHagimoto	0:0e0631af0305	883	If you need to extract a single channel or do some other sophisticated channel permutation, use
RyoheiHagimoto	0:0e0631af0305	884	mixChannels .
RyoheiHagimoto	0:0e0631af0305	885	@param src input multi-channel array.
RyoheiHagimoto	0:0e0631af0305	886	@param mvbegin output array; the number of arrays must match src.channels(); the arrays themselves are
RyoheiHagimoto	0:0e0631af0305	887	reallocated, if needed.
RyoheiHagimoto	0:0e0631af0305	888	@sa merge, mixChannels, cvtColor
RyoheiHagimoto	0:0e0631af0305	889	*/
RyoheiHagimoto	0:0e0631af0305	890	CV_EXPORTS void split(const Mat& src, Mat* mvbegin);
RyoheiHagimoto	0:0e0631af0305	891
RyoheiHagimoto	0:0e0631af0305	892	/** @overload
RyoheiHagimoto	0:0e0631af0305	893	@param m input multi-channel array.
RyoheiHagimoto	0:0e0631af0305	894	@param mv output vector of arrays; the arrays themselves are reallocated, if needed.
RyoheiHagimoto	0:0e0631af0305	895	*/
RyoheiHagimoto	0:0e0631af0305	896	CV_EXPORTS_W void split(InputArray m, OutputArrayOfArrays mv);
RyoheiHagimoto	0:0e0631af0305	897
RyoheiHagimoto	0:0e0631af0305	898	/** @brief Copies specified channels from input arrays to the specified channels of
RyoheiHagimoto	0:0e0631af0305	899	output arrays.
RyoheiHagimoto	0:0e0631af0305	900
RyoheiHagimoto	0:0e0631af0305	901	The function cv::mixChannels provides an advanced mechanism for shuffling image channels.
RyoheiHagimoto	0:0e0631af0305	902
RyoheiHagimoto	0:0e0631af0305	903	cv::split,cv::merge,cv::extractChannel,cv::insertChannel and some forms of cv::cvtColor are partial cases of cv::mixChannels.
RyoheiHagimoto	0:0e0631af0305	904
RyoheiHagimoto	0:0e0631af0305	905	In the example below, the code splits a 4-channel BGRA image into a 3-channel BGR (with B and R
RyoheiHagimoto	0:0e0631af0305	906	channels swapped) and a separate alpha-channel image:
RyoheiHagimoto	0:0e0631af0305	907	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	908	Mat bgra( 100, 100, CV_8UC4, Scalar(255,0,0,255) );
RyoheiHagimoto	0:0e0631af0305	909	Mat bgr( bgra.rows, bgra.cols, CV_8UC3 );
RyoheiHagimoto	0:0e0631af0305	910	Mat alpha( bgra.rows, bgra.cols, CV_8UC1 );
RyoheiHagimoto	0:0e0631af0305	911
RyoheiHagimoto	0:0e0631af0305	912	// forming an array of matrices is a quite efficient operation,
RyoheiHagimoto	0:0e0631af0305	913	// because the matrix data is not copied, only the headers
RyoheiHagimoto	0:0e0631af0305	914	Mat out[] = { bgr, alpha };
RyoheiHagimoto	0:0e0631af0305	915	// bgra[0] -> bgr[2], bgra[1] -> bgr[1],
RyoheiHagimoto	0:0e0631af0305	916	// bgra[2] -> bgr[0], bgra[3] -> alpha[0]
RyoheiHagimoto	0:0e0631af0305	917	int from_to[] = { 0,2, 1,1, 2,0, 3,3 };
RyoheiHagimoto	0:0e0631af0305	918	mixChannels( &bgra, 1, out, 2, from_to, 4 );
RyoheiHagimoto	0:0e0631af0305	919	@endcode
RyoheiHagimoto	0:0e0631af0305	920	@note Unlike many other new-style C++ functions in OpenCV (see the introduction section and
RyoheiHagimoto	0:0e0631af0305	921	Mat::create ), cv::mixChannels requires the output arrays to be pre-allocated before calling the
RyoheiHagimoto	0:0e0631af0305	922	function.
RyoheiHagimoto	0:0e0631af0305	923	@param src input array or vector of matrices; all of the matrices must have the same size and the
RyoheiHagimoto	0:0e0631af0305	924	same depth.
RyoheiHagimoto	0:0e0631af0305	925	@param nsrcs number of matrices in `src`.
RyoheiHagimoto	0:0e0631af0305	926	@param dst output array or vector of matrices; all the matrices must be allocated; their size and
RyoheiHagimoto	0:0e0631af0305	927	depth must be the same as in `src[0]`.
RyoheiHagimoto	0:0e0631af0305	928	@param ndsts number of matrices in `dst`.
RyoheiHagimoto	0:0e0631af0305	929	@param fromTo array of index pairs specifying which channels are copied and where; fromTo[k\*2] is
RyoheiHagimoto	0:0e0631af0305	930	a 0-based index of the input channel in src, fromTo[k\*2+1] is an index of the output channel in
RyoheiHagimoto	0:0e0631af0305	931	dst; the continuous channel numbering is used: the first input image channels are indexed from 0 to
RyoheiHagimoto	0:0e0631af0305	932	src[0].channels()-1, the second input image channels are indexed from src[0].channels() to
RyoheiHagimoto	0:0e0631af0305	933	src[0].channels() + src[1].channels()-1, and so on, the same scheme is used for the output image
RyoheiHagimoto	0:0e0631af0305	934	channels; as a special case, when fromTo[k\*2] is negative, the corresponding output channel is
RyoheiHagimoto	0:0e0631af0305	935	filled with zero .
RyoheiHagimoto	0:0e0631af0305	936	@param npairs number of index pairs in `fromTo`.
RyoheiHagimoto	0:0e0631af0305	937	@sa split, merge, extractChannel, insertChannel, cvtColor
RyoheiHagimoto	0:0e0631af0305	938	*/
RyoheiHagimoto	0:0e0631af0305	939	CV_EXPORTS void mixChannels(const Mat* src, size_t nsrcs, Mat* dst, size_t ndsts,
RyoheiHagimoto	0:0e0631af0305	940	const int* fromTo, size_t npairs);
RyoheiHagimoto	0:0e0631af0305	941
RyoheiHagimoto	0:0e0631af0305	942	/** @overload
RyoheiHagimoto	0:0e0631af0305	943	@param src input array or vector of matrices; all of the matrices must have the same size and the
RyoheiHagimoto	0:0e0631af0305	944	same depth.
RyoheiHagimoto	0:0e0631af0305	945	@param dst output array or vector of matrices; all the matrices must be allocated; their size and
RyoheiHagimoto	0:0e0631af0305	946	depth must be the same as in src[0].
RyoheiHagimoto	0:0e0631af0305	947	@param fromTo array of index pairs specifying which channels are copied and where; fromTo[k\*2] is
RyoheiHagimoto	0:0e0631af0305	948	a 0-based index of the input channel in src, fromTo[k\*2+1] is an index of the output channel in
RyoheiHagimoto	0:0e0631af0305	949	dst; the continuous channel numbering is used: the first input image channels are indexed from 0 to
RyoheiHagimoto	0:0e0631af0305	950	src[0].channels()-1, the second input image channels are indexed from src[0].channels() to
RyoheiHagimoto	0:0e0631af0305	951	src[0].channels() + src[1].channels()-1, and so on, the same scheme is used for the output image
RyoheiHagimoto	0:0e0631af0305	952	channels; as a special case, when fromTo[k\*2] is negative, the corresponding output channel is
RyoheiHagimoto	0:0e0631af0305	953	filled with zero .
RyoheiHagimoto	0:0e0631af0305	954	@param npairs number of index pairs in fromTo.
RyoheiHagimoto	0:0e0631af0305	955	*/
RyoheiHagimoto	0:0e0631af0305	956	CV_EXPORTS void mixChannels(InputArrayOfArrays src, InputOutputArrayOfArrays dst,
RyoheiHagimoto	0:0e0631af0305	957	const int* fromTo, size_t npairs);
RyoheiHagimoto	0:0e0631af0305	958
RyoheiHagimoto	0:0e0631af0305	959	/** @overload
RyoheiHagimoto	0:0e0631af0305	960	@param src input array or vector of matrices; all of the matrices must have the same size and the
RyoheiHagimoto	0:0e0631af0305	961	same depth.
RyoheiHagimoto	0:0e0631af0305	962	@param dst output array or vector of matrices; all the matrices must be allocated; their size and
RyoheiHagimoto	0:0e0631af0305	963	depth must be the same as in src[0].
RyoheiHagimoto	0:0e0631af0305	964	@param fromTo array of index pairs specifying which channels are copied and where; fromTo[k\*2] is
RyoheiHagimoto	0:0e0631af0305	965	a 0-based index of the input channel in src, fromTo[k\*2+1] is an index of the output channel in
RyoheiHagimoto	0:0e0631af0305	966	dst; the continuous channel numbering is used: the first input image channels are indexed from 0 to
RyoheiHagimoto	0:0e0631af0305	967	src[0].channels()-1, the second input image channels are indexed from src[0].channels() to
RyoheiHagimoto	0:0e0631af0305	968	src[0].channels() + src[1].channels()-1, and so on, the same scheme is used for the output image
RyoheiHagimoto	0:0e0631af0305	969	channels; as a special case, when fromTo[k\*2] is negative, the corresponding output channel is
RyoheiHagimoto	0:0e0631af0305	970	filled with zero .
RyoheiHagimoto	0:0e0631af0305	971	*/
RyoheiHagimoto	0:0e0631af0305	972	CV_EXPORTS_W void mixChannels(InputArrayOfArrays src, InputOutputArrayOfArrays dst,
RyoheiHagimoto	0:0e0631af0305	973	const std::vector<int>& fromTo);
RyoheiHagimoto	0:0e0631af0305	974
RyoheiHagimoto	0:0e0631af0305	975	/** @brief Extracts a single channel from src (coi is 0-based index)
RyoheiHagimoto	0:0e0631af0305	976	@param src input array
RyoheiHagimoto	0:0e0631af0305	977	@param dst output array
RyoheiHagimoto	0:0e0631af0305	978	@param coi index of channel to extract
RyoheiHagimoto	0:0e0631af0305	979	@sa mixChannels, split
RyoheiHagimoto	0:0e0631af0305	980	*/
RyoheiHagimoto	0:0e0631af0305	981	CV_EXPORTS_W void extractChannel(InputArray src, OutputArray dst, int coi);
RyoheiHagimoto	0:0e0631af0305	982
RyoheiHagimoto	0:0e0631af0305	983	/** @brief Inserts a single channel to dst (coi is 0-based index)
RyoheiHagimoto	0:0e0631af0305	984	@param src input array
RyoheiHagimoto	0:0e0631af0305	985	@param dst output array
RyoheiHagimoto	0:0e0631af0305	986	@param coi index of channel for insertion
RyoheiHagimoto	0:0e0631af0305	987	@sa mixChannels, merge
RyoheiHagimoto	0:0e0631af0305	988	*/
RyoheiHagimoto	0:0e0631af0305	989	CV_EXPORTS_W void insertChannel(InputArray src, InputOutputArray dst, int coi);
RyoheiHagimoto	0:0e0631af0305	990
RyoheiHagimoto	0:0e0631af0305	991	/** @brief Flips a 2D array around vertical, horizontal, or both axes.
RyoheiHagimoto	0:0e0631af0305	992
RyoheiHagimoto	0:0e0631af0305	993	The function cv::flip flips the array in one of three different ways (row
RyoheiHagimoto	0:0e0631af0305	994	and column indices are 0-based):
RyoheiHagimoto	0:0e0631af0305	995	\f[\texttt{dst} _{ij} =
RyoheiHagimoto	0:0e0631af0305	996	\left\{
RyoheiHagimoto	0:0e0631af0305	997	\begin{array}{l l}
RyoheiHagimoto	0:0e0631af0305	998	\texttt{src} _{\texttt{src.rows}-i-1,j} & if\; \texttt{flipCode} = 0 \\
RyoheiHagimoto	0:0e0631af0305	999	\texttt{src} _{i, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} > 0 \\
RyoheiHagimoto	0:0e0631af0305	1000	\texttt{src} _{ \texttt{src.rows} -i-1, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} < 0 \\
RyoheiHagimoto	0:0e0631af0305	1001	\end{array}
RyoheiHagimoto	0:0e0631af0305	1002	\right.\f]
RyoheiHagimoto	0:0e0631af0305	1003	The example scenarios of using the function are the following:
RyoheiHagimoto	0:0e0631af0305	1004	* Vertical flipping of the image (flipCode == 0) to switch between
RyoheiHagimoto	0:0e0631af0305	1005	top-left and bottom-left image origin. This is a typical operation
RyoheiHagimoto	0:0e0631af0305	1006	in video processing on Microsoft Windows\* OS.
RyoheiHagimoto	0:0e0631af0305	1007	* Horizontal flipping of the image with the subsequent horizontal
RyoheiHagimoto	0:0e0631af0305	1008	shift and absolute difference calculation to check for a
RyoheiHagimoto	0:0e0631af0305	1009	vertical-axis symmetry (flipCode \> 0).
RyoheiHagimoto	0:0e0631af0305	1010	* Simultaneous horizontal and vertical flipping of the image with
RyoheiHagimoto	0:0e0631af0305	1011	the subsequent shift and absolute difference calculation to check
RyoheiHagimoto	0:0e0631af0305	1012	for a central symmetry (flipCode \< 0).
RyoheiHagimoto	0:0e0631af0305	1013	* Reversing the order of point arrays (flipCode \> 0 or
RyoheiHagimoto	0:0e0631af0305	1014	flipCode == 0).
RyoheiHagimoto	0:0e0631af0305	1015	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1016	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1017	@param flipCode a flag to specify how to flip the array; 0 means
RyoheiHagimoto	0:0e0631af0305	1018	flipping around the x-axis and positive value (for example, 1) means
RyoheiHagimoto	0:0e0631af0305	1019	flipping around y-axis. Negative value (for example, -1) means flipping
RyoheiHagimoto	0:0e0631af0305	1020	around both axes.
RyoheiHagimoto	0:0e0631af0305	1021	@sa transpose , repeat , completeSymm
RyoheiHagimoto	0:0e0631af0305	1022	*/
RyoheiHagimoto	0:0e0631af0305	1023	CV_EXPORTS_W void flip(InputArray src, OutputArray dst, int flipCode);
RyoheiHagimoto	0:0e0631af0305	1024
RyoheiHagimoto	0:0e0631af0305	1025	enum RotateFlags {
RyoheiHagimoto	0:0e0631af0305	1026	ROTATE_90_CLOCKWISE = 0, //Rotate 90 degrees clockwise
RyoheiHagimoto	0:0e0631af0305	1027	ROTATE_180 = 1, //Rotate 180 degrees clockwise
RyoheiHagimoto	0:0e0631af0305	1028	ROTATE_90_COUNTERCLOCKWISE = 2, //Rotate 270 degrees clockwise
RyoheiHagimoto	0:0e0631af0305	1029	};
RyoheiHagimoto	0:0e0631af0305	1030	/** @brief Rotates a 2D array in multiples of 90 degrees.
RyoheiHagimoto	0:0e0631af0305	1031	The function rotate rotates the array in one of three different ways:
RyoheiHagimoto	0:0e0631af0305	1032	* Rotate by 90 degrees clockwise (rotateCode = ROTATE_90).
RyoheiHagimoto	0:0e0631af0305	1033	* Rotate by 180 degrees clockwise (rotateCode = ROTATE_180).
RyoheiHagimoto	0:0e0631af0305	1034	* Rotate by 270 degrees clockwise (rotateCode = ROTATE_270).
RyoheiHagimoto	0:0e0631af0305	1035	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1036	@param dst output array of the same type as src. The size is the same with ROTATE_180,
RyoheiHagimoto	0:0e0631af0305	1037	and the rows and cols are switched for ROTATE_90 and ROTATE_270.
RyoheiHagimoto	0:0e0631af0305	1038	@param rotateCode an enum to specify how to rotate the array; see the enum RotateFlags
RyoheiHagimoto	0:0e0631af0305	1039	@sa transpose , repeat , completeSymm, flip, RotateFlags
RyoheiHagimoto	0:0e0631af0305	1040	*/
RyoheiHagimoto	0:0e0631af0305	1041	CV_EXPORTS_W void rotate(InputArray src, OutputArray dst, int rotateCode);
RyoheiHagimoto	0:0e0631af0305	1042
RyoheiHagimoto	0:0e0631af0305	1043	/** @brief Fills the output array with repeated copies of the input array.
RyoheiHagimoto	0:0e0631af0305	1044
RyoheiHagimoto	0:0e0631af0305	1045	The function cv::repeat duplicates the input array one or more times along each of the two axes:
RyoheiHagimoto	0:0e0631af0305	1046	\f[\texttt{dst} _{ij}= \texttt{src} _{i\mod src.rows, \; j\mod src.cols }\f]
RyoheiHagimoto	0:0e0631af0305	1047	The second variant of the function is more convenient to use with @ref MatrixExpressions.
RyoheiHagimoto	0:0e0631af0305	1048	@param src input array to replicate.
RyoheiHagimoto	0:0e0631af0305	1049	@param ny Flag to specify how many times the `src` is repeated along the
RyoheiHagimoto	0:0e0631af0305	1050	vertical axis.
RyoheiHagimoto	0:0e0631af0305	1051	@param nx Flag to specify how many times the `src` is repeated along the
RyoheiHagimoto	0:0e0631af0305	1052	horizontal axis.
RyoheiHagimoto	0:0e0631af0305	1053	@param dst output array of the same type as `src`.
RyoheiHagimoto	0:0e0631af0305	1054	@sa cv::reduce
RyoheiHagimoto	0:0e0631af0305	1055	*/
RyoheiHagimoto	0:0e0631af0305	1056	CV_EXPORTS_W void repeat(InputArray src, int ny, int nx, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1057
RyoheiHagimoto	0:0e0631af0305	1058	/** @overload
RyoheiHagimoto	0:0e0631af0305	1059	@param src input array to replicate.
RyoheiHagimoto	0:0e0631af0305	1060	@param ny Flag to specify how many times the `src` is repeated along the
RyoheiHagimoto	0:0e0631af0305	1061	vertical axis.
RyoheiHagimoto	0:0e0631af0305	1062	@param nx Flag to specify how many times the `src` is repeated along the
RyoheiHagimoto	0:0e0631af0305	1063	horizontal axis.
RyoheiHagimoto	0:0e0631af0305	1064	*/
RyoheiHagimoto	0:0e0631af0305	1065	CV_EXPORTS Mat repeat(const Mat& src, int ny, int nx);
RyoheiHagimoto	0:0e0631af0305	1066
RyoheiHagimoto	0:0e0631af0305	1067	/** @brief Applies horizontal concatenation to given matrices.
RyoheiHagimoto	0:0e0631af0305	1068
RyoheiHagimoto	0:0e0631af0305	1069	The function horizontally concatenates two or more cv::Mat matrices (with the same number of rows).
RyoheiHagimoto	0:0e0631af0305	1070	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1071	cv::Mat matArray[] = { cv::Mat(4, 1, CV_8UC1, cv::Scalar(1)),
RyoheiHagimoto	0:0e0631af0305	1072	cv::Mat(4, 1, CV_8UC1, cv::Scalar(2)),
RyoheiHagimoto	0:0e0631af0305	1073	cv::Mat(4, 1, CV_8UC1, cv::Scalar(3)),};
RyoheiHagimoto	0:0e0631af0305	1074
RyoheiHagimoto	0:0e0631af0305	1075	cv::Mat out;
RyoheiHagimoto	0:0e0631af0305	1076	cv::hconcat( matArray, 3, out );
RyoheiHagimoto	0:0e0631af0305	1077	//out:
RyoheiHagimoto	0:0e0631af0305	1078	//[1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1079	// 1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1080	// 1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1081	// 1, 2, 3]
RyoheiHagimoto	0:0e0631af0305	1082	@endcode
RyoheiHagimoto	0:0e0631af0305	1083	@param src input array or vector of matrices. all of the matrices must have the same number of rows and the same depth.
RyoheiHagimoto	0:0e0631af0305	1084	@param nsrc number of matrices in src.
RyoheiHagimoto	0:0e0631af0305	1085	@param dst output array. It has the same number of rows and depth as the src, and the sum of cols of the src.
RyoheiHagimoto	0:0e0631af0305	1086	@sa cv::vconcat(const Mat*, size_t, OutputArray), @sa cv::vconcat(InputArrayOfArrays, OutputArray) and @sa cv::vconcat(InputArray, InputArray, OutputArray)
RyoheiHagimoto	0:0e0631af0305	1087	*/
RyoheiHagimoto	0:0e0631af0305	1088	CV_EXPORTS void hconcat(const Mat* src, size_t nsrc, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1089	/** @overload
RyoheiHagimoto	0:0e0631af0305	1090	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1091	cv::Mat_<float> A = (cv::Mat_<float>(3, 2) << 1, 4,
RyoheiHagimoto	0:0e0631af0305	1092	2, 5,
RyoheiHagimoto	0:0e0631af0305	1093	3, 6);
RyoheiHagimoto	0:0e0631af0305	1094	cv::Mat_<float> B = (cv::Mat_<float>(3, 2) << 7, 10,
RyoheiHagimoto	0:0e0631af0305	1095	8, 11,
RyoheiHagimoto	0:0e0631af0305	1096	9, 12);
RyoheiHagimoto	0:0e0631af0305	1097
RyoheiHagimoto	0:0e0631af0305	1098	cv::Mat C;
RyoheiHagimoto	0:0e0631af0305	1099	cv::hconcat(A, B, C);
RyoheiHagimoto	0:0e0631af0305	1100	//C:
RyoheiHagimoto	0:0e0631af0305	1101	//[1, 4, 7, 10;
RyoheiHagimoto	0:0e0631af0305	1102	// 2, 5, 8, 11;
RyoheiHagimoto	0:0e0631af0305	1103	// 3, 6, 9, 12]
RyoheiHagimoto	0:0e0631af0305	1104	@endcode
RyoheiHagimoto	0:0e0631af0305	1105	@param src1 first input array to be considered for horizontal concatenation.
RyoheiHagimoto	0:0e0631af0305	1106	@param src2 second input array to be considered for horizontal concatenation.
RyoheiHagimoto	0:0e0631af0305	1107	@param dst output array. It has the same number of rows and depth as the src1 and src2, and the sum of cols of the src1 and src2.
RyoheiHagimoto	0:0e0631af0305	1108	*/
RyoheiHagimoto	0:0e0631af0305	1109	CV_EXPORTS void hconcat(InputArray src1, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1110	/** @overload
RyoheiHagimoto	0:0e0631af0305	1111	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1112	std::vector<cv::Mat> matrices = { cv::Mat(4, 1, CV_8UC1, cv::Scalar(1)),
RyoheiHagimoto	0:0e0631af0305	1113	cv::Mat(4, 1, CV_8UC1, cv::Scalar(2)),
RyoheiHagimoto	0:0e0631af0305	1114	cv::Mat(4, 1, CV_8UC1, cv::Scalar(3)),};
RyoheiHagimoto	0:0e0631af0305	1115
RyoheiHagimoto	0:0e0631af0305	1116	cv::Mat out;
RyoheiHagimoto	0:0e0631af0305	1117	cv::hconcat( matrices, out );
RyoheiHagimoto	0:0e0631af0305	1118	//out:
RyoheiHagimoto	0:0e0631af0305	1119	//[1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1120	// 1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1121	// 1, 2, 3;
RyoheiHagimoto	0:0e0631af0305	1122	// 1, 2, 3]
RyoheiHagimoto	0:0e0631af0305	1123	@endcode
RyoheiHagimoto	0:0e0631af0305	1124	@param src input array or vector of matrices. all of the matrices must have the same number of rows and the same depth.
RyoheiHagimoto	0:0e0631af0305	1125	@param dst output array. It has the same number of rows and depth as the src, and the sum of cols of the src.
RyoheiHagimoto	0:0e0631af0305	1126	same depth.
RyoheiHagimoto	0:0e0631af0305	1127	*/
RyoheiHagimoto	0:0e0631af0305	1128	CV_EXPORTS_W void hconcat(InputArrayOfArrays src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1129
RyoheiHagimoto	0:0e0631af0305	1130	/** @brief Applies vertical concatenation to given matrices.
RyoheiHagimoto	0:0e0631af0305	1131
RyoheiHagimoto	0:0e0631af0305	1132	The function vertically concatenates two or more cv::Mat matrices (with the same number of cols).
RyoheiHagimoto	0:0e0631af0305	1133	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1134	cv::Mat matArray[] = { cv::Mat(1, 4, CV_8UC1, cv::Scalar(1)),
RyoheiHagimoto	0:0e0631af0305	1135	cv::Mat(1, 4, CV_8UC1, cv::Scalar(2)),
RyoheiHagimoto	0:0e0631af0305	1136	cv::Mat(1, 4, CV_8UC1, cv::Scalar(3)),};
RyoheiHagimoto	0:0e0631af0305	1137
RyoheiHagimoto	0:0e0631af0305	1138	cv::Mat out;
RyoheiHagimoto	0:0e0631af0305	1139	cv::vconcat( matArray, 3, out );
RyoheiHagimoto	0:0e0631af0305	1140	//out:
RyoheiHagimoto	0:0e0631af0305	1141	//[1, 1, 1, 1;
RyoheiHagimoto	0:0e0631af0305	1142	// 2, 2, 2, 2;
RyoheiHagimoto	0:0e0631af0305	1143	// 3, 3, 3, 3]
RyoheiHagimoto	0:0e0631af0305	1144	@endcode
RyoheiHagimoto	0:0e0631af0305	1145	@param src input array or vector of matrices. all of the matrices must have the same number of cols and the same depth.
RyoheiHagimoto	0:0e0631af0305	1146	@param nsrc number of matrices in src.
RyoheiHagimoto	0:0e0631af0305	1147	@param dst output array. It has the same number of cols and depth as the src, and the sum of rows of the src.
RyoheiHagimoto	0:0e0631af0305	1148	@sa cv::hconcat(const Mat*, size_t, OutputArray), @sa cv::hconcat(InputArrayOfArrays, OutputArray) and @sa cv::hconcat(InputArray, InputArray, OutputArray)
RyoheiHagimoto	0:0e0631af0305	1149	*/
RyoheiHagimoto	0:0e0631af0305	1150	CV_EXPORTS void vconcat(const Mat* src, size_t nsrc, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1151	/** @overload
RyoheiHagimoto	0:0e0631af0305	1152	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1153	cv::Mat_<float> A = (cv::Mat_<float>(3, 2) << 1, 7,
RyoheiHagimoto	0:0e0631af0305	1154	2, 8,
RyoheiHagimoto	0:0e0631af0305	1155	3, 9);
RyoheiHagimoto	0:0e0631af0305	1156	cv::Mat_<float> B = (cv::Mat_<float>(3, 2) << 4, 10,
RyoheiHagimoto	0:0e0631af0305	1157	5, 11,
RyoheiHagimoto	0:0e0631af0305	1158	6, 12);
RyoheiHagimoto	0:0e0631af0305	1159
RyoheiHagimoto	0:0e0631af0305	1160	cv::Mat C;
RyoheiHagimoto	0:0e0631af0305	1161	cv::vconcat(A, B, C);
RyoheiHagimoto	0:0e0631af0305	1162	//C:
RyoheiHagimoto	0:0e0631af0305	1163	//[1, 7;
RyoheiHagimoto	0:0e0631af0305	1164	// 2, 8;
RyoheiHagimoto	0:0e0631af0305	1165	// 3, 9;
RyoheiHagimoto	0:0e0631af0305	1166	// 4, 10;
RyoheiHagimoto	0:0e0631af0305	1167	// 5, 11;
RyoheiHagimoto	0:0e0631af0305	1168	// 6, 12]
RyoheiHagimoto	0:0e0631af0305	1169	@endcode
RyoheiHagimoto	0:0e0631af0305	1170	@param src1 first input array to be considered for vertical concatenation.
RyoheiHagimoto	0:0e0631af0305	1171	@param src2 second input array to be considered for vertical concatenation.
RyoheiHagimoto	0:0e0631af0305	1172	@param dst output array. It has the same number of cols and depth as the src1 and src2, and the sum of rows of the src1 and src2.
RyoheiHagimoto	0:0e0631af0305	1173	*/
RyoheiHagimoto	0:0e0631af0305	1174	CV_EXPORTS void vconcat(InputArray src1, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1175	/** @overload
RyoheiHagimoto	0:0e0631af0305	1176	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1177	std::vector<cv::Mat> matrices = { cv::Mat(1, 4, CV_8UC1, cv::Scalar(1)),
RyoheiHagimoto	0:0e0631af0305	1178	cv::Mat(1, 4, CV_8UC1, cv::Scalar(2)),
RyoheiHagimoto	0:0e0631af0305	1179	cv::Mat(1, 4, CV_8UC1, cv::Scalar(3)),};
RyoheiHagimoto	0:0e0631af0305	1180
RyoheiHagimoto	0:0e0631af0305	1181	cv::Mat out;
RyoheiHagimoto	0:0e0631af0305	1182	cv::vconcat( matrices, out );
RyoheiHagimoto	0:0e0631af0305	1183	//out:
RyoheiHagimoto	0:0e0631af0305	1184	//[1, 1, 1, 1;
RyoheiHagimoto	0:0e0631af0305	1185	// 2, 2, 2, 2;
RyoheiHagimoto	0:0e0631af0305	1186	// 3, 3, 3, 3]
RyoheiHagimoto	0:0e0631af0305	1187	@endcode
RyoheiHagimoto	0:0e0631af0305	1188	@param src input array or vector of matrices. all of the matrices must have the same number of cols and the same depth
RyoheiHagimoto	0:0e0631af0305	1189	@param dst output array. It has the same number of cols and depth as the src, and the sum of rows of the src.
RyoheiHagimoto	0:0e0631af0305	1190	same depth.
RyoheiHagimoto	0:0e0631af0305	1191	*/
RyoheiHagimoto	0:0e0631af0305	1192	CV_EXPORTS_W void vconcat(InputArrayOfArrays src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1193
RyoheiHagimoto	0:0e0631af0305	1194	/** @brief computes bitwise conjunction of the two arrays (dst = src1 & src2)
RyoheiHagimoto	0:0e0631af0305	1195	Calculates the per-element bit-wise conjunction of two arrays or an
RyoheiHagimoto	0:0e0631af0305	1196	array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1197
RyoheiHagimoto	0:0e0631af0305	1198	The function cv::bitwise_and calculates the per-element bit-wise logical conjunction for:
RyoheiHagimoto	0:0e0631af0305	1199	* Two arrays when src1 and src2 have the same size:
RyoheiHagimoto	0:0e0631af0305	1200	\f[\texttt{dst} (I) = \texttt{src1} (I) \wedge \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1201	* An array and a scalar when src2 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1202	the same number of elements as `src1.channels()`:
RyoheiHagimoto	0:0e0631af0305	1203	\f[\texttt{dst} (I) = \texttt{src1} (I) \wedge \texttt{src2} \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1204	* A scalar and an array when src1 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1205	the same number of elements as `src2.channels()`:
RyoheiHagimoto	0:0e0631af0305	1206	\f[\texttt{dst} (I) = \texttt{src1} \wedge \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1207	In case of floating-point arrays, their machine-specific bit
RyoheiHagimoto	0:0e0631af0305	1208	representations (usually IEEE754-compliant) are used for the operation.
RyoheiHagimoto	0:0e0631af0305	1209	In case of multi-channel arrays, each channel is processed
RyoheiHagimoto	0:0e0631af0305	1210	independently. In the second and third cases above, the scalar is first
RyoheiHagimoto	0:0e0631af0305	1211	converted to the array type.
RyoheiHagimoto	0:0e0631af0305	1212	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1213	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1214	@param dst output array that has the same size and type as the input
RyoheiHagimoto	0:0e0631af0305	1215	arrays.
RyoheiHagimoto	0:0e0631af0305	1216	@param mask optional operation mask, 8-bit single channel array, that
RyoheiHagimoto	0:0e0631af0305	1217	specifies elements of the output array to be changed.
RyoheiHagimoto	0:0e0631af0305	1218	*/
RyoheiHagimoto	0:0e0631af0305	1219	CV_EXPORTS_W void bitwise_and(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	1220	OutputArray dst, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	1221
RyoheiHagimoto	0:0e0631af0305	1222	/** @brief Calculates the per-element bit-wise disjunction of two arrays or an
RyoheiHagimoto	0:0e0631af0305	1223	array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1224
RyoheiHagimoto	0:0e0631af0305	1225	The function cv::bitwise_or calculates the per-element bit-wise logical disjunction for:
RyoheiHagimoto	0:0e0631af0305	1226	* Two arrays when src1 and src2 have the same size:
RyoheiHagimoto	0:0e0631af0305	1227	\f[\texttt{dst} (I) = \texttt{src1} (I) \vee \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1228	* An array and a scalar when src2 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1229	the same number of elements as `src1.channels()`:
RyoheiHagimoto	0:0e0631af0305	1230	\f[\texttt{dst} (I) = \texttt{src1} (I) \vee \texttt{src2} \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1231	* A scalar and an array when src1 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1232	the same number of elements as `src2.channels()`:
RyoheiHagimoto	0:0e0631af0305	1233	\f[\texttt{dst} (I) = \texttt{src1} \vee \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1234	In case of floating-point arrays, their machine-specific bit
RyoheiHagimoto	0:0e0631af0305	1235	representations (usually IEEE754-compliant) are used for the operation.
RyoheiHagimoto	0:0e0631af0305	1236	In case of multi-channel arrays, each channel is processed
RyoheiHagimoto	0:0e0631af0305	1237	independently. In the second and third cases above, the scalar is first
RyoheiHagimoto	0:0e0631af0305	1238	converted to the array type.
RyoheiHagimoto	0:0e0631af0305	1239	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1240	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1241	@param dst output array that has the same size and type as the input
RyoheiHagimoto	0:0e0631af0305	1242	arrays.
RyoheiHagimoto	0:0e0631af0305	1243	@param mask optional operation mask, 8-bit single channel array, that
RyoheiHagimoto	0:0e0631af0305	1244	specifies elements of the output array to be changed.
RyoheiHagimoto	0:0e0631af0305	1245	*/
RyoheiHagimoto	0:0e0631af0305	1246	CV_EXPORTS_W void bitwise_or(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	1247	OutputArray dst, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	1248
RyoheiHagimoto	0:0e0631af0305	1249	/** @brief Calculates the per-element bit-wise "exclusive or" operation on two
RyoheiHagimoto	0:0e0631af0305	1250	arrays or an array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1251
RyoheiHagimoto	0:0e0631af0305	1252	The function cv::bitwise_xor calculates the per-element bit-wise logical "exclusive-or"
RyoheiHagimoto	0:0e0631af0305	1253	operation for:
RyoheiHagimoto	0:0e0631af0305	1254	* Two arrays when src1 and src2 have the same size:
RyoheiHagimoto	0:0e0631af0305	1255	\f[\texttt{dst} (I) = \texttt{src1} (I) \oplus \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1256	* An array and a scalar when src2 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1257	the same number of elements as `src1.channels()`:
RyoheiHagimoto	0:0e0631af0305	1258	\f[\texttt{dst} (I) = \texttt{src1} (I) \oplus \texttt{src2} \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1259	* A scalar and an array when src1 is constructed from Scalar or has
RyoheiHagimoto	0:0e0631af0305	1260	the same number of elements as `src2.channels()`:
RyoheiHagimoto	0:0e0631af0305	1261	\f[\texttt{dst} (I) = \texttt{src1} \oplus \texttt{src2} (I) \quad \texttt{if mask} (I) \ne0\f]
RyoheiHagimoto	0:0e0631af0305	1262	In case of floating-point arrays, their machine-specific bit
RyoheiHagimoto	0:0e0631af0305	1263	representations (usually IEEE754-compliant) are used for the operation.
RyoheiHagimoto	0:0e0631af0305	1264	In case of multi-channel arrays, each channel is processed
RyoheiHagimoto	0:0e0631af0305	1265	independently. In the 2nd and 3rd cases above, the scalar is first
RyoheiHagimoto	0:0e0631af0305	1266	converted to the array type.
RyoheiHagimoto	0:0e0631af0305	1267	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1268	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1269	@param dst output array that has the same size and type as the input
RyoheiHagimoto	0:0e0631af0305	1270	arrays.
RyoheiHagimoto	0:0e0631af0305	1271	@param mask optional operation mask, 8-bit single channel array, that
RyoheiHagimoto	0:0e0631af0305	1272	specifies elements of the output array to be changed.
RyoheiHagimoto	0:0e0631af0305	1273	*/
RyoheiHagimoto	0:0e0631af0305	1274	CV_EXPORTS_W void bitwise_xor(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	1275	OutputArray dst, InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	1276
RyoheiHagimoto	0:0e0631af0305	1277	/** @brief Inverts every bit of an array.
RyoheiHagimoto	0:0e0631af0305	1278
RyoheiHagimoto	0:0e0631af0305	1279	The function cv::bitwise_not calculates per-element bit-wise inversion of the input
RyoheiHagimoto	0:0e0631af0305	1280	array:
RyoheiHagimoto	0:0e0631af0305	1281	\f[\texttt{dst} (I) = \neg \texttt{src} (I)\f]
RyoheiHagimoto	0:0e0631af0305	1282	In case of a floating-point input array, its machine-specific bit
RyoheiHagimoto	0:0e0631af0305	1283	representation (usually IEEE754-compliant) is used for the operation. In
RyoheiHagimoto	0:0e0631af0305	1284	case of multi-channel arrays, each channel is processed independently.
RyoheiHagimoto	0:0e0631af0305	1285	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1286	@param dst output array that has the same size and type as the input
RyoheiHagimoto	0:0e0631af0305	1287	array.
RyoheiHagimoto	0:0e0631af0305	1288	@param mask optional operation mask, 8-bit single channel array, that
RyoheiHagimoto	0:0e0631af0305	1289	specifies elements of the output array to be changed.
RyoheiHagimoto	0:0e0631af0305	1290	*/
RyoheiHagimoto	0:0e0631af0305	1291	CV_EXPORTS_W void bitwise_not(InputArray src, OutputArray dst,
RyoheiHagimoto	0:0e0631af0305	1292	InputArray mask = noArray());
RyoheiHagimoto	0:0e0631af0305	1293
RyoheiHagimoto	0:0e0631af0305	1294	/** @brief Calculates the per-element absolute difference between two arrays or between an array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1295
RyoheiHagimoto	0:0e0631af0305	1296	The function cv::absdiff calculates:
RyoheiHagimoto	0:0e0631af0305	1297	* Absolute difference between two arrays when they have the same
RyoheiHagimoto	0:0e0631af0305	1298	size and type:
RyoheiHagimoto	0:0e0631af0305	1299	\f[\texttt{dst}(I) = \texttt{saturate} (\| \texttt{src1}(I) - \texttt{src2}(I)\|)\f]
RyoheiHagimoto	0:0e0631af0305	1300	* Absolute difference between an array and a scalar when the second
RyoheiHagimoto	0:0e0631af0305	1301	array is constructed from Scalar or has as many elements as the
RyoheiHagimoto	0:0e0631af0305	1302	number of channels in `src1`:
RyoheiHagimoto	0:0e0631af0305	1303	\f[\texttt{dst}(I) = \texttt{saturate} (\| \texttt{src1}(I) - \texttt{src2} \|)\f]
RyoheiHagimoto	0:0e0631af0305	1304	* Absolute difference between a scalar and an array when the first
RyoheiHagimoto	0:0e0631af0305	1305	array is constructed from Scalar or has as many elements as the
RyoheiHagimoto	0:0e0631af0305	1306	number of channels in `src2`:
RyoheiHagimoto	0:0e0631af0305	1307	\f[\texttt{dst}(I) = \texttt{saturate} (\| \texttt{src1} - \texttt{src2}(I) \|)\f]
RyoheiHagimoto	0:0e0631af0305	1308	where I is a multi-dimensional index of array elements. In case of
RyoheiHagimoto	0:0e0631af0305	1309	multi-channel arrays, each channel is processed independently.
RyoheiHagimoto	0:0e0631af0305	1310	@note Saturation is not applied when the arrays have the depth CV_32S.
RyoheiHagimoto	0:0e0631af0305	1311	You may even get a negative value in the case of overflow.
RyoheiHagimoto	0:0e0631af0305	1312	@param src1 first input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1313	@param src2 second input array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1314	@param dst output array that has the same size and type as input arrays.
RyoheiHagimoto	0:0e0631af0305	1315	@sa cv::abs(const Mat&)
RyoheiHagimoto	0:0e0631af0305	1316	*/
RyoheiHagimoto	0:0e0631af0305	1317	CV_EXPORTS_W void absdiff(InputArray src1, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1318
RyoheiHagimoto	0:0e0631af0305	1319	/** @brief Checks if array elements lie between the elements of two other arrays.
RyoheiHagimoto	0:0e0631af0305	1320
RyoheiHagimoto	0:0e0631af0305	1321	The function checks the range as follows:
RyoheiHagimoto	0:0e0631af0305	1322	- For every element of a single-channel input array:
RyoheiHagimoto	0:0e0631af0305	1323	\f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0\f]
RyoheiHagimoto	0:0e0631af0305	1324	- For two-channel arrays:
RyoheiHagimoto	0:0e0631af0305	1325	\f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0 \land \texttt{lowerb} (I)_1 \leq \texttt{src} (I)_1 \leq \texttt{upperb} (I)_1\f]
RyoheiHagimoto	0:0e0631af0305	1326	- and so forth.
RyoheiHagimoto	0:0e0631af0305	1327
RyoheiHagimoto	0:0e0631af0305	1328	That is, dst (I) is set to 255 (all 1 -bits) if src (I) is within the
RyoheiHagimoto	0:0e0631af0305	1329	specified 1D, 2D, 3D, ... box and 0 otherwise.
RyoheiHagimoto	0:0e0631af0305	1330
RyoheiHagimoto	0:0e0631af0305	1331	When the lower and/or upper boundary parameters are scalars, the indexes
RyoheiHagimoto	0:0e0631af0305	1332	(I) at lowerb and upperb in the above formulas should be omitted.
RyoheiHagimoto	0:0e0631af0305	1333	@param src first input array.
RyoheiHagimoto	0:0e0631af0305	1334	@param lowerb inclusive lower boundary array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1335	@param upperb inclusive upper boundary array or a scalar.
RyoheiHagimoto	0:0e0631af0305	1336	@param dst output array of the same size as src and CV_8U type.
RyoheiHagimoto	0:0e0631af0305	1337	*/
RyoheiHagimoto	0:0e0631af0305	1338	CV_EXPORTS_W void inRange(InputArray src, InputArray lowerb,
RyoheiHagimoto	0:0e0631af0305	1339	InputArray upperb, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1340
RyoheiHagimoto	0:0e0631af0305	1341	/** @brief Performs the per-element comparison of two arrays or an array and scalar value.
RyoheiHagimoto	0:0e0631af0305	1342
RyoheiHagimoto	0:0e0631af0305	1343	The function compares:
RyoheiHagimoto	0:0e0631af0305	1344	* Elements of two arrays when src1 and src2 have the same size:
RyoheiHagimoto	0:0e0631af0305	1345	\f[\texttt{dst} (I) = \texttt{src1} (I) \,\texttt{cmpop}\, \texttt{src2} (I)\f]
RyoheiHagimoto	0:0e0631af0305	1346	* Elements of src1 with a scalar src2 when src2 is constructed from
RyoheiHagimoto	0:0e0631af0305	1347	Scalar or has a single element:
RyoheiHagimoto	0:0e0631af0305	1348	\f[\texttt{dst} (I) = \texttt{src1}(I) \,\texttt{cmpop}\, \texttt{src2}\f]
RyoheiHagimoto	0:0e0631af0305	1349	* src1 with elements of src2 when src1 is constructed from Scalar or
RyoheiHagimoto	0:0e0631af0305	1350	has a single element:
RyoheiHagimoto	0:0e0631af0305	1351	\f[\texttt{dst} (I) = \texttt{src1} \,\texttt{cmpop}\, \texttt{src2} (I)\f]
RyoheiHagimoto	0:0e0631af0305	1352	When the comparison result is true, the corresponding element of output
RyoheiHagimoto	0:0e0631af0305	1353	array is set to 255. The comparison operations can be replaced with the
RyoheiHagimoto	0:0e0631af0305	1354	equivalent matrix expressions:
RyoheiHagimoto	0:0e0631af0305	1355	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1356	Mat dst1 = src1 >= src2;
RyoheiHagimoto	0:0e0631af0305	1357	Mat dst2 = src1 < 8;
RyoheiHagimoto	0:0e0631af0305	1358	...
RyoheiHagimoto	0:0e0631af0305	1359	@endcode
RyoheiHagimoto	0:0e0631af0305	1360	@param src1 first input array or a scalar; when it is an array, it must have a single channel.
RyoheiHagimoto	0:0e0631af0305	1361	@param src2 second input array or a scalar; when it is an array, it must have a single channel.
RyoheiHagimoto	0:0e0631af0305	1362	@param dst output array of type ref CV_8U that has the same size and the same number of channels as
RyoheiHagimoto	0:0e0631af0305	1363	the input arrays.
RyoheiHagimoto	0:0e0631af0305	1364	@param cmpop a flag, that specifies correspondence between the arrays (cv::CmpTypes)
RyoheiHagimoto	0:0e0631af0305	1365	@sa checkRange, min, max, threshold
RyoheiHagimoto	0:0e0631af0305	1366	*/
RyoheiHagimoto	0:0e0631af0305	1367	CV_EXPORTS_W void compare(InputArray src1, InputArray src2, OutputArray dst, int cmpop);
RyoheiHagimoto	0:0e0631af0305	1368
RyoheiHagimoto	0:0e0631af0305	1369	/** @brief Calculates per-element minimum of two arrays or an array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1370
RyoheiHagimoto	0:0e0631af0305	1371	The function cv::min calculates the per-element minimum of two arrays:
RyoheiHagimoto	0:0e0631af0305	1372	\f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{src2} (I))\f]
RyoheiHagimoto	0:0e0631af0305	1373	or array and a scalar:
RyoheiHagimoto	0:0e0631af0305	1374	\f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{value} )\f]
RyoheiHagimoto	0:0e0631af0305	1375	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	1376	@param src2 second input array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	1377	@param dst output array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	1378	@sa max, compare, inRange, minMaxLoc
RyoheiHagimoto	0:0e0631af0305	1379	*/
RyoheiHagimoto	0:0e0631af0305	1380	CV_EXPORTS_W void min(InputArray src1, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1381	/** @overload
RyoheiHagimoto	0:0e0631af0305	1382	needed to avoid conflicts with const _Tp& std::min(const _Tp&, const _Tp&, _Compare)
RyoheiHagimoto	0:0e0631af0305	1383	*/
RyoheiHagimoto	0:0e0631af0305	1384	CV_EXPORTS void min(const Mat& src1, const Mat& src2, Mat& dst);
RyoheiHagimoto	0:0e0631af0305	1385	/** @overload
RyoheiHagimoto	0:0e0631af0305	1386	needed to avoid conflicts with const _Tp& std::min(const _Tp&, const _Tp&, _Compare)
RyoheiHagimoto	0:0e0631af0305	1387	*/
RyoheiHagimoto	0:0e0631af0305	1388	CV_EXPORTS void min(const UMat& src1, const UMat& src2, UMat& dst);
RyoheiHagimoto	0:0e0631af0305	1389
RyoheiHagimoto	0:0e0631af0305	1390	/** @brief Calculates per-element maximum of two arrays or an array and a scalar.
RyoheiHagimoto	0:0e0631af0305	1391
RyoheiHagimoto	0:0e0631af0305	1392	The function cv::max calculates the per-element maximum of two arrays:
RyoheiHagimoto	0:0e0631af0305	1393	\f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{src2} (I))\f]
RyoheiHagimoto	0:0e0631af0305	1394	or array and a scalar:
RyoheiHagimoto	0:0e0631af0305	1395	\f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{value} )\f]
RyoheiHagimoto	0:0e0631af0305	1396	@param src1 first input array.
RyoheiHagimoto	0:0e0631af0305	1397	@param src2 second input array of the same size and type as src1 .
RyoheiHagimoto	0:0e0631af0305	1398	@param dst output array of the same size and type as src1.
RyoheiHagimoto	0:0e0631af0305	1399	@sa min, compare, inRange, minMaxLoc, @ref MatrixExpressions
RyoheiHagimoto	0:0e0631af0305	1400	*/
RyoheiHagimoto	0:0e0631af0305	1401	CV_EXPORTS_W void max(InputArray src1, InputArray src2, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1402	/** @overload
RyoheiHagimoto	0:0e0631af0305	1403	needed to avoid conflicts with const _Tp& std::min(const _Tp&, const _Tp&, _Compare)
RyoheiHagimoto	0:0e0631af0305	1404	*/
RyoheiHagimoto	0:0e0631af0305	1405	CV_EXPORTS void max(const Mat& src1, const Mat& src2, Mat& dst);
RyoheiHagimoto	0:0e0631af0305	1406	/** @overload
RyoheiHagimoto	0:0e0631af0305	1407	needed to avoid conflicts with const _Tp& std::min(const _Tp&, const _Tp&, _Compare)
RyoheiHagimoto	0:0e0631af0305	1408	*/
RyoheiHagimoto	0:0e0631af0305	1409	CV_EXPORTS void max(const UMat& src1, const UMat& src2, UMat& dst);
RyoheiHagimoto	0:0e0631af0305	1410
RyoheiHagimoto	0:0e0631af0305	1411	/** @brief Calculates a square root of array elements.
RyoheiHagimoto	0:0e0631af0305	1412
RyoheiHagimoto	0:0e0631af0305	1413	The function cv::sqrt calculates a square root of each input array element.
RyoheiHagimoto	0:0e0631af0305	1414	In case of multi-channel arrays, each channel is processed
RyoheiHagimoto	0:0e0631af0305	1415	independently. The accuracy is approximately the same as of the built-in
RyoheiHagimoto	0:0e0631af0305	1416	std::sqrt .
RyoheiHagimoto	0:0e0631af0305	1417	@param src input floating-point array.
RyoheiHagimoto	0:0e0631af0305	1418	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1419	*/
RyoheiHagimoto	0:0e0631af0305	1420	CV_EXPORTS_W void sqrt(InputArray src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1421
RyoheiHagimoto	0:0e0631af0305	1422	/** @brief Raises every array element to a power.
RyoheiHagimoto	0:0e0631af0305	1423
RyoheiHagimoto	0:0e0631af0305	1424	The function cv::pow raises every element of the input array to power :
RyoheiHagimoto	0:0e0631af0305	1425	\f[\texttt{dst} (I) = \fork{\texttt{src}(I)^{power}}{if $\texttt{power}$ is integer}{\|\texttt{src}(I)\|^{power}}{otherwise}\f]
RyoheiHagimoto	0:0e0631af0305	1426
RyoheiHagimoto	0:0e0631af0305	1427	So, for a non-integer power exponent, the absolute values of input array
RyoheiHagimoto	0:0e0631af0305	1428	elements are used. However, it is possible to get true values for
RyoheiHagimoto	0:0e0631af0305	1429	negative values using some extra operations. In the example below,
RyoheiHagimoto	0:0e0631af0305	1430	computing the 5th root of array src shows:
RyoheiHagimoto	0:0e0631af0305	1431	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1432	Mat mask = src < 0;
RyoheiHagimoto	0:0e0631af0305	1433	pow(src, 1./5, dst);
RyoheiHagimoto	0:0e0631af0305	1434	subtract(Scalar::all(0), dst, dst, mask);
RyoheiHagimoto	0:0e0631af0305	1435	@endcode
RyoheiHagimoto	0:0e0631af0305	1436	For some values of power, such as integer values, 0.5 and -0.5,
RyoheiHagimoto	0:0e0631af0305	1437	specialized faster algorithms are used.
RyoheiHagimoto	0:0e0631af0305	1438
RyoheiHagimoto	0:0e0631af0305	1439	Special values (NaN, Inf) are not handled.
RyoheiHagimoto	0:0e0631af0305	1440	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1441	@param power exponent of power.
RyoheiHagimoto	0:0e0631af0305	1442	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1443	@sa sqrt, exp, log, cartToPolar, polarToCart
RyoheiHagimoto	0:0e0631af0305	1444	*/
RyoheiHagimoto	0:0e0631af0305	1445	CV_EXPORTS_W void pow(InputArray src, double power, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1446
RyoheiHagimoto	0:0e0631af0305	1447	/** @brief Calculates the exponent of every array element.
RyoheiHagimoto	0:0e0631af0305	1448
RyoheiHagimoto	0:0e0631af0305	1449	The function cv::exp calculates the exponent of every element of the input
RyoheiHagimoto	0:0e0631af0305	1450	array:
RyoheiHagimoto	0:0e0631af0305	1451	\f[\texttt{dst} [I] = e^{ src(I) }\f]
RyoheiHagimoto	0:0e0631af0305	1452
RyoheiHagimoto	0:0e0631af0305	1453	The maximum relative error is about 7e-6 for single-precision input and
RyoheiHagimoto	0:0e0631af0305	1454	less than 1e-10 for double-precision input. Currently, the function
RyoheiHagimoto	0:0e0631af0305	1455	converts denormalized values to zeros on output. Special values (NaN,
RyoheiHagimoto	0:0e0631af0305	1456	Inf) are not handled.
RyoheiHagimoto	0:0e0631af0305	1457	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1458	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1459	@sa log , cartToPolar , polarToCart , phase , pow , sqrt , magnitude
RyoheiHagimoto	0:0e0631af0305	1460	*/
RyoheiHagimoto	0:0e0631af0305	1461	CV_EXPORTS_W void exp(InputArray src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1462
RyoheiHagimoto	0:0e0631af0305	1463	/** @brief Calculates the natural logarithm of every array element.
RyoheiHagimoto	0:0e0631af0305	1464
RyoheiHagimoto	0:0e0631af0305	1465	The function cv::log calculates the natural logarithm of every element of the input array:
RyoheiHagimoto	0:0e0631af0305	1466	\f[\texttt{dst} (I) = \log (\texttt{src}(I)) \f]
RyoheiHagimoto	0:0e0631af0305	1467
RyoheiHagimoto	0:0e0631af0305	1468	Output on zero, negative and special (NaN, Inf) values is undefined.
RyoheiHagimoto	0:0e0631af0305	1469
RyoheiHagimoto	0:0e0631af0305	1470	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1471	@param dst output array of the same size and type as src .
RyoheiHagimoto	0:0e0631af0305	1472	@sa exp, cartToPolar, polarToCart, phase, pow, sqrt, magnitude
RyoheiHagimoto	0:0e0631af0305	1473	*/
RyoheiHagimoto	0:0e0631af0305	1474	CV_EXPORTS_W void log(InputArray src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1475
RyoheiHagimoto	0:0e0631af0305	1476	/** @brief Calculates x and y coordinates of 2D vectors from their magnitude and angle.
RyoheiHagimoto	0:0e0631af0305	1477
RyoheiHagimoto	0:0e0631af0305	1478	The function cv::polarToCart calculates the Cartesian coordinates of each 2D
RyoheiHagimoto	0:0e0631af0305	1479	vector represented by the corresponding elements of magnitude and angle:
RyoheiHagimoto	0:0e0631af0305	1480	\f[\begin{array}{l} \texttt{x} (I) = \texttt{magnitude} (I) \cos ( \texttt{angle} (I)) \\ \texttt{y} (I) = \texttt{magnitude} (I) \sin ( \texttt{angle} (I)) \\ \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	1481
RyoheiHagimoto	0:0e0631af0305	1482	The relative accuracy of the estimated coordinates is about 1e-6.
RyoheiHagimoto	0:0e0631af0305	1483	@param magnitude input floating-point array of magnitudes of 2D vectors;
RyoheiHagimoto	0:0e0631af0305	1484	it can be an empty matrix (=Mat()), in this case, the function assumes
RyoheiHagimoto	0:0e0631af0305	1485	that all the magnitudes are =1; if it is not empty, it must have the
RyoheiHagimoto	0:0e0631af0305	1486	same size and type as angle.
RyoheiHagimoto	0:0e0631af0305	1487	@param angle input floating-point array of angles of 2D vectors.
RyoheiHagimoto	0:0e0631af0305	1488	@param x output array of x-coordinates of 2D vectors; it has the same
RyoheiHagimoto	0:0e0631af0305	1489	size and type as angle.
RyoheiHagimoto	0:0e0631af0305	1490	@param y output array of y-coordinates of 2D vectors; it has the same
RyoheiHagimoto	0:0e0631af0305	1491	size and type as angle.
RyoheiHagimoto	0:0e0631af0305	1492	@param angleInDegrees when true, the input angles are measured in
RyoheiHagimoto	0:0e0631af0305	1493	degrees, otherwise, they are measured in radians.
RyoheiHagimoto	0:0e0631af0305	1494	@sa cartToPolar, magnitude, phase, exp, log, pow, sqrt
RyoheiHagimoto	0:0e0631af0305	1495	*/
RyoheiHagimoto	0:0e0631af0305	1496	CV_EXPORTS_W void polarToCart(InputArray magnitude, InputArray angle,
RyoheiHagimoto	0:0e0631af0305	1497	OutputArray x, OutputArray y, bool angleInDegrees = false);
RyoheiHagimoto	0:0e0631af0305	1498
RyoheiHagimoto	0:0e0631af0305	1499	/** @brief Calculates the magnitude and angle of 2D vectors.
RyoheiHagimoto	0:0e0631af0305	1500
RyoheiHagimoto	0:0e0631af0305	1501	The function cv::cartToPolar calculates either the magnitude, angle, or both
RyoheiHagimoto	0:0e0631af0305	1502	for every 2D vector (x(I),y(I)):
RyoheiHagimoto	0:0e0631af0305	1503	\f[\begin{array}{l} \texttt{magnitude} (I)= \sqrt{\texttt{x}(I)^2+\texttt{y}(I)^2} , \\ \texttt{angle} (I)= \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))[ \cdot180 / \pi ] \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	1504
RyoheiHagimoto	0:0e0631af0305	1505	The angles are calculated with accuracy about 0.3 degrees. For the point
RyoheiHagimoto	0:0e0631af0305	1506	(0,0), the angle is set to 0.
RyoheiHagimoto	0:0e0631af0305	1507	@param x array of x-coordinates; this must be a single-precision or
RyoheiHagimoto	0:0e0631af0305	1508	double-precision floating-point array.
RyoheiHagimoto	0:0e0631af0305	1509	@param y array of y-coordinates, that must have the same size and same type as x.
RyoheiHagimoto	0:0e0631af0305	1510	@param magnitude output array of magnitudes of the same size and type as x.
RyoheiHagimoto	0:0e0631af0305	1511	@param angle output array of angles that has the same size and type as
RyoheiHagimoto	0:0e0631af0305	1512	x; the angles are measured in radians (from 0 to 2\*Pi) or in degrees (0 to 360 degrees).
RyoheiHagimoto	0:0e0631af0305	1513	@param angleInDegrees a flag, indicating whether the angles are measured
RyoheiHagimoto	0:0e0631af0305	1514	in radians (which is by default), or in degrees.
RyoheiHagimoto	0:0e0631af0305	1515	@sa Sobel, Scharr
RyoheiHagimoto	0:0e0631af0305	1516	*/
RyoheiHagimoto	0:0e0631af0305	1517	CV_EXPORTS_W void cartToPolar(InputArray x, InputArray y,
RyoheiHagimoto	0:0e0631af0305	1518	OutputArray magnitude, OutputArray angle,
RyoheiHagimoto	0:0e0631af0305	1519	bool angleInDegrees = false);
RyoheiHagimoto	0:0e0631af0305	1520
RyoheiHagimoto	0:0e0631af0305	1521	/** @brief Calculates the rotation angle of 2D vectors.
RyoheiHagimoto	0:0e0631af0305	1522
RyoheiHagimoto	0:0e0631af0305	1523	The function cv::phase calculates the rotation angle of each 2D vector that
RyoheiHagimoto	0:0e0631af0305	1524	is formed from the corresponding elements of x and y :
RyoheiHagimoto	0:0e0631af0305	1525	\f[\texttt{angle} (I) = \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))\f]
RyoheiHagimoto	0:0e0631af0305	1526
RyoheiHagimoto	0:0e0631af0305	1527	The angle estimation accuracy is about 0.3 degrees. When x(I)=y(I)=0 ,
RyoheiHagimoto	0:0e0631af0305	1528	the corresponding angle(I) is set to 0.
RyoheiHagimoto	0:0e0631af0305	1529	@param x input floating-point array of x-coordinates of 2D vectors.
RyoheiHagimoto	0:0e0631af0305	1530	@param y input array of y-coordinates of 2D vectors; it must have the
RyoheiHagimoto	0:0e0631af0305	1531	same size and the same type as x.
RyoheiHagimoto	0:0e0631af0305	1532	@param angle output array of vector angles; it has the same size and
RyoheiHagimoto	0:0e0631af0305	1533	same type as x .
RyoheiHagimoto	0:0e0631af0305	1534	@param angleInDegrees when true, the function calculates the angle in
RyoheiHagimoto	0:0e0631af0305	1535	degrees, otherwise, they are measured in radians.
RyoheiHagimoto	0:0e0631af0305	1536	*/
RyoheiHagimoto	0:0e0631af0305	1537	CV_EXPORTS_W void phase(InputArray x, InputArray y, OutputArray angle,
RyoheiHagimoto	0:0e0631af0305	1538	bool angleInDegrees = false);
RyoheiHagimoto	0:0e0631af0305	1539
RyoheiHagimoto	0:0e0631af0305	1540	/** @brief Calculates the magnitude of 2D vectors.
RyoheiHagimoto	0:0e0631af0305	1541
RyoheiHagimoto	0:0e0631af0305	1542	The function cv::magnitude calculates the magnitude of 2D vectors formed
RyoheiHagimoto	0:0e0631af0305	1543	from the corresponding elements of x and y arrays:
RyoheiHagimoto	0:0e0631af0305	1544	\f[\texttt{dst} (I) = \sqrt{\texttt{x}(I)^2 + \texttt{y}(I)^2}\f]
RyoheiHagimoto	0:0e0631af0305	1545	@param x floating-point array of x-coordinates of the vectors.
RyoheiHagimoto	0:0e0631af0305	1546	@param y floating-point array of y-coordinates of the vectors; it must
RyoheiHagimoto	0:0e0631af0305	1547	have the same size as x.
RyoheiHagimoto	0:0e0631af0305	1548	@param magnitude output array of the same size and type as x.
RyoheiHagimoto	0:0e0631af0305	1549	@sa cartToPolar, polarToCart, phase, sqrt
RyoheiHagimoto	0:0e0631af0305	1550	*/
RyoheiHagimoto	0:0e0631af0305	1551	CV_EXPORTS_W void magnitude(InputArray x, InputArray y, OutputArray magnitude);
RyoheiHagimoto	0:0e0631af0305	1552
RyoheiHagimoto	0:0e0631af0305	1553	/** @brief Checks every element of an input array for invalid values.
RyoheiHagimoto	0:0e0631af0305	1554
RyoheiHagimoto	0:0e0631af0305	1555	The function cv::checkRange checks that every array element is neither NaN nor infinite. When minVal \>
RyoheiHagimoto	0:0e0631af0305	1556	-DBL_MAX and maxVal \< DBL_MAX, the function also checks that each value is between minVal and
RyoheiHagimoto	0:0e0631af0305	1557	maxVal. In case of multi-channel arrays, each channel is processed independently. If some values
RyoheiHagimoto	0:0e0631af0305	1558	are out of range, position of the first outlier is stored in pos (when pos != NULL). Then, the
RyoheiHagimoto	0:0e0631af0305	1559	function either returns false (when quiet=true) or throws an exception.
RyoheiHagimoto	0:0e0631af0305	1560	@param a input array.
RyoheiHagimoto	0:0e0631af0305	1561	@param quiet a flag, indicating whether the functions quietly return false when the array elements
RyoheiHagimoto	0:0e0631af0305	1562	are out of range or they throw an exception.
RyoheiHagimoto	0:0e0631af0305	1563	@param pos optional output parameter, when not NULL, must be a pointer to array of src.dims
RyoheiHagimoto	0:0e0631af0305	1564	elements.
RyoheiHagimoto	0:0e0631af0305	1565	@param minVal inclusive lower boundary of valid values range.
RyoheiHagimoto	0:0e0631af0305	1566	@param maxVal exclusive upper boundary of valid values range.
RyoheiHagimoto	0:0e0631af0305	1567	*/
RyoheiHagimoto	0:0e0631af0305	1568	CV_EXPORTS_W bool checkRange(InputArray a, bool quiet = true, CV_OUT Point* pos = 0,
RyoheiHagimoto	0:0e0631af0305	1569	double minVal = -DBL_MAX, double maxVal = DBL_MAX);
RyoheiHagimoto	0:0e0631af0305	1570
RyoheiHagimoto	0:0e0631af0305	1571	/** @brief converts NaN's to the given number
RyoheiHagimoto	0:0e0631af0305	1572	*/
RyoheiHagimoto	0:0e0631af0305	1573	CV_EXPORTS_W void patchNaNs(InputOutputArray a, double val = 0);
RyoheiHagimoto	0:0e0631af0305	1574
RyoheiHagimoto	0:0e0631af0305	1575	/** @brief Performs generalized matrix multiplication.
RyoheiHagimoto	0:0e0631af0305	1576
RyoheiHagimoto	0:0e0631af0305	1577	The function cv::gemm performs generalized matrix multiplication similar to the
RyoheiHagimoto	0:0e0631af0305	1578	gemm functions in BLAS level 3. For example,
RyoheiHagimoto	0:0e0631af0305	1579	`gemm(src1, src2, alpha, src3, beta, dst, GEMM_1_T + GEMM_3_T)`
RyoheiHagimoto	0:0e0631af0305	1580	corresponds to
RyoheiHagimoto	0:0e0631af0305	1581	\f[\texttt{dst} = \texttt{alpha} \cdot \texttt{src1} ^T \cdot \texttt{src2} + \texttt{beta} \cdot \texttt{src3} ^T\f]
RyoheiHagimoto	0:0e0631af0305	1582
RyoheiHagimoto	0:0e0631af0305	1583	In case of complex (two-channel) data, performed a complex matrix
RyoheiHagimoto	0:0e0631af0305	1584	multiplication.
RyoheiHagimoto	0:0e0631af0305	1585
RyoheiHagimoto	0:0e0631af0305	1586	The function can be replaced with a matrix expression. For example, the
RyoheiHagimoto	0:0e0631af0305	1587	above call can be replaced with:
RyoheiHagimoto	0:0e0631af0305	1588	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	1589	dst = alphasrc1.t()src2 + beta*src3.t();
RyoheiHagimoto	0:0e0631af0305	1590	@endcode
RyoheiHagimoto	0:0e0631af0305	1591	@param src1 first multiplied input matrix that could be real(CV_32FC1,
RyoheiHagimoto	0:0e0631af0305	1592	CV_64FC1) or complex(CV_32FC2, CV_64FC2).
RyoheiHagimoto	0:0e0631af0305	1593	@param src2 second multiplied input matrix of the same type as src1.
RyoheiHagimoto	0:0e0631af0305	1594	@param alpha weight of the matrix product.
RyoheiHagimoto	0:0e0631af0305	1595	@param src3 third optional delta matrix added to the matrix product; it
RyoheiHagimoto	0:0e0631af0305	1596	should have the same type as src1 and src2.
RyoheiHagimoto	0:0e0631af0305	1597	@param beta weight of src3.
RyoheiHagimoto	0:0e0631af0305	1598	@param dst output matrix; it has the proper size and the same type as
RyoheiHagimoto	0:0e0631af0305	1599	input matrices.
RyoheiHagimoto	0:0e0631af0305	1600	@param flags operation flags (cv::GemmFlags)
RyoheiHagimoto	0:0e0631af0305	1601	@sa mulTransposed , transform
RyoheiHagimoto	0:0e0631af0305	1602	*/
RyoheiHagimoto	0:0e0631af0305	1603	CV_EXPORTS_W void gemm(InputArray src1, InputArray src2, double alpha,
RyoheiHagimoto	0:0e0631af0305	1604	InputArray src3, double beta, OutputArray dst, int flags = 0);
RyoheiHagimoto	0:0e0631af0305	1605
RyoheiHagimoto	0:0e0631af0305	1606	/** @brief Calculates the product of a matrix and its transposition.
RyoheiHagimoto	0:0e0631af0305	1607
RyoheiHagimoto	0:0e0631af0305	1608	The function cv::mulTransposed calculates the product of src and its
RyoheiHagimoto	0:0e0631af0305	1609	transposition:
RyoheiHagimoto	0:0e0631af0305	1610	\f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} )^T ( \texttt{src} - \texttt{delta} )\f]
RyoheiHagimoto	0:0e0631af0305	1611	if aTa=true , and
RyoheiHagimoto	0:0e0631af0305	1612	\f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} ) ( \texttt{src} - \texttt{delta} )^T\f]
RyoheiHagimoto	0:0e0631af0305	1613	otherwise. The function is used to calculate the covariance matrix. With
RyoheiHagimoto	0:0e0631af0305	1614	zero delta, it can be used as a faster substitute for general matrix
RyoheiHagimoto	0:0e0631af0305	1615	product A\*B when B=A'
RyoheiHagimoto	0:0e0631af0305	1616	@param src input single-channel matrix. Note that unlike gemm, the
RyoheiHagimoto	0:0e0631af0305	1617	function can multiply not only floating-point matrices.
RyoheiHagimoto	0:0e0631af0305	1618	@param dst output square matrix.
RyoheiHagimoto	0:0e0631af0305	1619	@param aTa Flag specifying the multiplication ordering. See the
RyoheiHagimoto	0:0e0631af0305	1620	description below.
RyoheiHagimoto	0:0e0631af0305	1621	@param delta Optional delta matrix subtracted from src before the
RyoheiHagimoto	0:0e0631af0305	1622	multiplication. When the matrix is empty ( delta=noArray() ), it is
RyoheiHagimoto	0:0e0631af0305	1623	assumed to be zero, that is, nothing is subtracted. If it has the same
RyoheiHagimoto	0:0e0631af0305	1624	size as src , it is simply subtracted. Otherwise, it is "repeated" (see
RyoheiHagimoto	0:0e0631af0305	1625	repeat ) to cover the full src and then subtracted. Type of the delta
RyoheiHagimoto	0:0e0631af0305	1626	matrix, when it is not empty, must be the same as the type of created
RyoheiHagimoto	0:0e0631af0305	1627	output matrix. See the dtype parameter description below.
RyoheiHagimoto	0:0e0631af0305	1628	@param scale Optional scale factor for the matrix product.
RyoheiHagimoto	0:0e0631af0305	1629	@param dtype Optional type of the output matrix. When it is negative,
RyoheiHagimoto	0:0e0631af0305	1630	the output matrix will have the same type as src . Otherwise, it will be
RyoheiHagimoto	0:0e0631af0305	1631	type=CV_MAT_DEPTH(dtype) that should be either CV_32F or CV_64F .
RyoheiHagimoto	0:0e0631af0305	1632	@sa calcCovarMatrix, gemm, repeat, reduce
RyoheiHagimoto	0:0e0631af0305	1633	*/
RyoheiHagimoto	0:0e0631af0305	1634	CV_EXPORTS_W void mulTransposed( InputArray src, OutputArray dst, bool aTa,
RyoheiHagimoto	0:0e0631af0305	1635	InputArray delta = noArray(),
RyoheiHagimoto	0:0e0631af0305	1636	double scale = 1, int dtype = -1 );
RyoheiHagimoto	0:0e0631af0305	1637
RyoheiHagimoto	0:0e0631af0305	1638	/** @brief Transposes a matrix.
RyoheiHagimoto	0:0e0631af0305	1639
RyoheiHagimoto	0:0e0631af0305	1640	The function cv::transpose transposes the matrix src :
RyoheiHagimoto	0:0e0631af0305	1641	\f[\texttt{dst} (i,j) = \texttt{src} (j,i)\f]
RyoheiHagimoto	0:0e0631af0305	1642	@note No complex conjugation is done in case of a complex matrix. It it
RyoheiHagimoto	0:0e0631af0305	1643	should be done separately if needed.
RyoheiHagimoto	0:0e0631af0305	1644	@param src input array.
RyoheiHagimoto	0:0e0631af0305	1645	@param dst output array of the same type as src.
RyoheiHagimoto	0:0e0631af0305	1646	*/
RyoheiHagimoto	0:0e0631af0305	1647	CV_EXPORTS_W void transpose(InputArray src, OutputArray dst);
RyoheiHagimoto	0:0e0631af0305	1648
RyoheiHagimoto	0:0e0631af0305	1649	/** @brief Performs the matrix transformation of every array element.
RyoheiHagimoto	0:0e0631af0305	1650
RyoheiHagimoto	0:0e0631af0305	1651	The function cv::transform performs the matrix transformation of every
RyoheiHagimoto	0:0e0631af0305	1652	element of the array src and stores the results in dst :
RyoheiHagimoto	0:0e0631af0305	1653	\f[\texttt{dst} (I) = \texttt{m} \cdot \texttt{src} (I)\f]
RyoheiHagimoto	0:0e0631af0305	1654	(when m.cols=src.channels() ), or
RyoheiHagimoto	0:0e0631af0305	1655	\f[\texttt{dst} (I) = \texttt{m} \cdot [ \texttt{src} (I); 1]\f]
RyoheiHagimoto	0:0e0631af0305	1656	(when m.cols=src.channels()+1 )
RyoheiHagimoto	0:0e0631af0305	1657
RyoheiHagimoto	0:0e0631af0305	1658	Every element of the N -channel array src is interpreted as N -element
RyoheiHagimoto	0:0e0631af0305	1659	vector that is transformed using the M x N or M x (N+1) matrix m to
RyoheiHagimoto	0:0e0631af0305	1660	M-element vector - the corresponding element of the output array dst .
RyoheiHagimoto	0:0e0631af0305	1661
RyoheiHagimoto	0:0e0631af0305	1662	The function may be used for geometrical transformation of
RyoheiHagimoto	0:0e0631af0305	1663	N -dimensional points, arbitrary linear color space transformation (such
RyoheiHagimoto	0:0e0631af0305	1664	as various kinds of RGB to YUV transforms), shuffling the image
RyoheiHagimoto	0:0e0631af0305	1665	channels, and so forth.
RyoheiHagimoto	0:0e0631af0305	1666	@param src input array that must have as many channels (1 to 4) as
RyoheiHagimoto	0:0e0631af0305	1667	m.cols or m.cols-1.
RyoheiHagimoto	0:0e0631af0305	1668	@param dst output array of the same size and depth as src; it has as
RyoheiHagimoto	0:0e0631af0305	1669	many channels as m.rows.
RyoheiHagimoto	0:0e0631af0305	1670	@param m transformation 2x2 or 2x3 floating-point matrix.
RyoheiHagimoto	0:0e0631af0305	1671	@sa perspectiveTransform, getAffineTransform, estimateAffine2D, warpAffine, warpPerspective
RyoheiHagimoto	0:0e0631af0305	1672	*/
RyoheiHagimoto	0:0e0631af0305	1673	CV_EXPORTS_W void transform(InputArray src, OutputArray dst, InputArray m );
RyoheiHagimoto	0:0e0631af0305	1674
RyoheiHagimoto	0:0e0631af0305	1675	/** @brief Performs the perspective matrix transformation of vectors.
RyoheiHagimoto	0:0e0631af0305	1676
RyoheiHagimoto	0:0e0631af0305	1677	The function cv::perspectiveTransform transforms every element of src by
RyoheiHagimoto	0:0e0631af0305	1678	treating it as a 2D or 3D vector, in the following way:
RyoheiHagimoto	0:0e0631af0305	1679	\f[(x, y, z) \rightarrow (x'/w, y'/w, z'/w)\f]
RyoheiHagimoto	0:0e0631af0305	1680	where
RyoheiHagimoto	0:0e0631af0305	1681	\f[(x', y', z', w') = \texttt{mat} \cdot \begin{bmatrix} x & y & z & 1 \end{bmatrix}\f]
RyoheiHagimoto	0:0e0631af0305	1682	and
RyoheiHagimoto	0:0e0631af0305	1683	\f[w = \fork{w'}{if $w' \ne 0$}{\infty}{otherwise}\f]
RyoheiHagimoto	0:0e0631af0305	1684
RyoheiHagimoto	0:0e0631af0305	1685	Here a 3D vector transformation is shown. In case of a 2D vector
RyoheiHagimoto	0:0e0631af0305	1686	transformation, the z component is omitted.
RyoheiHagimoto	0:0e0631af0305	1687
RyoheiHagimoto	0:0e0631af0305	1688	@note The function transforms a sparse set of 2D or 3D vectors. If you
RyoheiHagimoto	0:0e0631af0305	1689	want to transform an image using perspective transformation, use
RyoheiHagimoto	0:0e0631af0305	1690	warpPerspective . If you have an inverse problem, that is, you want to
RyoheiHagimoto	0:0e0631af0305	1691	compute the most probable perspective transformation out of several
RyoheiHagimoto	0:0e0631af0305	1692	pairs of corresponding points, you can use getPerspectiveTransform or
RyoheiHagimoto	0:0e0631af0305	1693	findHomography .
RyoheiHagimoto	0:0e0631af0305	1694	@param src input two-channel or three-channel floating-point array; each
RyoheiHagimoto	0:0e0631af0305	1695	element is a 2D/3D vector to be transformed.
RyoheiHagimoto	0:0e0631af0305	1696	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1697	@param m 3x3 or 4x4 floating-point transformation matrix.
RyoheiHagimoto	0:0e0631af0305	1698	@sa transform, warpPerspective, getPerspectiveTransform, findHomography
RyoheiHagimoto	0:0e0631af0305	1699	*/
RyoheiHagimoto	0:0e0631af0305	1700	CV_EXPORTS_W void perspectiveTransform(InputArray src, OutputArray dst, InputArray m );
RyoheiHagimoto	0:0e0631af0305	1701
RyoheiHagimoto	0:0e0631af0305	1702	/** @brief Copies the lower or the upper half of a square matrix to another half.
RyoheiHagimoto	0:0e0631af0305	1703
RyoheiHagimoto	0:0e0631af0305	1704	The function cv::completeSymm copies the lower half of a square matrix to
RyoheiHagimoto	0:0e0631af0305	1705	its another half. The matrix diagonal remains unchanged:
RyoheiHagimoto	0:0e0631af0305	1706	* \f$\texttt{mtx}_{ij}=\texttt{mtx}_{ji}\f$ for \f$i > j\f$ if
RyoheiHagimoto	0:0e0631af0305	1707	lowerToUpper=false
RyoheiHagimoto	0:0e0631af0305	1708	* \f$\texttt{mtx}_{ij}=\texttt{mtx}_{ji}\f$ for \f$i < j\f$ if
RyoheiHagimoto	0:0e0631af0305	1709	lowerToUpper=true
RyoheiHagimoto	0:0e0631af0305	1710	@param mtx input-output floating-point square matrix.
RyoheiHagimoto	0:0e0631af0305	1711	@param lowerToUpper operation flag; if true, the lower half is copied to
RyoheiHagimoto	0:0e0631af0305	1712	the upper half. Otherwise, the upper half is copied to the lower half.
RyoheiHagimoto	0:0e0631af0305	1713	@sa flip, transpose
RyoheiHagimoto	0:0e0631af0305	1714	*/
RyoheiHagimoto	0:0e0631af0305	1715	CV_EXPORTS_W void completeSymm(InputOutputArray mtx, bool lowerToUpper = false);
RyoheiHagimoto	0:0e0631af0305	1716
RyoheiHagimoto	0:0e0631af0305	1717	/** @brief Initializes a scaled identity matrix.
RyoheiHagimoto	0:0e0631af0305	1718
RyoheiHagimoto	0:0e0631af0305	1719	The function cv::setIdentity initializes a scaled identity matrix:
RyoheiHagimoto	0:0e0631af0305	1720	\f[\texttt{mtx} (i,j)= \fork{\texttt{value}}{ if $i=j$}{0}{otherwise}\f]
RyoheiHagimoto	0:0e0631af0305	1721
RyoheiHagimoto	0:0e0631af0305	1722	The function can also be emulated using the matrix initializers and the
RyoheiHagimoto	0:0e0631af0305	1723	matrix expressions:
RyoheiHagimoto	0:0e0631af0305	1724	@code
RyoheiHagimoto	0:0e0631af0305	1725	Mat A = Mat::eye(4, 3, CV_32F)*5;
RyoheiHagimoto	0:0e0631af0305	1726	// A will be set to [[5, 0, 0], [0, 5, 0], [0, 0, 5], [0, 0, 0]]
RyoheiHagimoto	0:0e0631af0305	1727	@endcode
RyoheiHagimoto	0:0e0631af0305	1728	@param mtx matrix to initialize (not necessarily square).
RyoheiHagimoto	0:0e0631af0305	1729	@param s value to assign to diagonal elements.
RyoheiHagimoto	0:0e0631af0305	1730	@sa Mat::zeros, Mat::ones, Mat::setTo, Mat::operator=
RyoheiHagimoto	0:0e0631af0305	1731	*/
RyoheiHagimoto	0:0e0631af0305	1732	CV_EXPORTS_W void setIdentity(InputOutputArray mtx, const Scalar& s = Scalar(1));
RyoheiHagimoto	0:0e0631af0305	1733
RyoheiHagimoto	0:0e0631af0305	1734	/** @brief Returns the determinant of a square floating-point matrix.
RyoheiHagimoto	0:0e0631af0305	1735
RyoheiHagimoto	0:0e0631af0305	1736	The function cv::determinant calculates and returns the determinant of the
RyoheiHagimoto	0:0e0631af0305	1737	specified matrix. For small matrices ( mtx.cols=mtx.rows\<=3 ), the
RyoheiHagimoto	0:0e0631af0305	1738	direct method is used. For larger matrices, the function uses LU
RyoheiHagimoto	0:0e0631af0305	1739	factorization with partial pivoting.
RyoheiHagimoto	0:0e0631af0305	1740
RyoheiHagimoto	0:0e0631af0305	1741	For symmetric positively-determined matrices, it is also possible to use
RyoheiHagimoto	0:0e0631af0305	1742	eigen decomposition to calculate the determinant.
RyoheiHagimoto	0:0e0631af0305	1743	@param mtx input matrix that must have CV_32FC1 or CV_64FC1 type and
RyoheiHagimoto	0:0e0631af0305	1744	square size.
RyoheiHagimoto	0:0e0631af0305	1745	@sa trace, invert, solve, eigen, @ref MatrixExpressions
RyoheiHagimoto	0:0e0631af0305	1746	*/
RyoheiHagimoto	0:0e0631af0305	1747	CV_EXPORTS_W double determinant(InputArray mtx);
RyoheiHagimoto	0:0e0631af0305	1748
RyoheiHagimoto	0:0e0631af0305	1749	/** @brief Returns the trace of a matrix.
RyoheiHagimoto	0:0e0631af0305	1750
RyoheiHagimoto	0:0e0631af0305	1751	The function cv::trace returns the sum of the diagonal elements of the
RyoheiHagimoto	0:0e0631af0305	1752	matrix mtx .
RyoheiHagimoto	0:0e0631af0305	1753	\f[\mathrm{tr} ( \texttt{mtx} ) = \sum _i \texttt{mtx} (i,i)\f]
RyoheiHagimoto	0:0e0631af0305	1754	@param mtx input matrix.
RyoheiHagimoto	0:0e0631af0305	1755	*/
RyoheiHagimoto	0:0e0631af0305	1756	CV_EXPORTS_W Scalar trace(InputArray mtx);
RyoheiHagimoto	0:0e0631af0305	1757
RyoheiHagimoto	0:0e0631af0305	1758	/** @brief Finds the inverse or pseudo-inverse of a matrix.
RyoheiHagimoto	0:0e0631af0305	1759
RyoheiHagimoto	0:0e0631af0305	1760	The function cv::invert inverts the matrix src and stores the result in dst
RyoheiHagimoto	0:0e0631af0305	1761	. When the matrix src is singular or non-square, the function calculates
RyoheiHagimoto	0:0e0631af0305	1762	the pseudo-inverse matrix (the dst matrix) so that norm(src\*dst - I) is
RyoheiHagimoto	0:0e0631af0305	1763	minimal, where I is an identity matrix.
RyoheiHagimoto	0:0e0631af0305	1764
RyoheiHagimoto	0:0e0631af0305	1765	In case of the DECOMP_LU method, the function returns non-zero value if
RyoheiHagimoto	0:0e0631af0305	1766	the inverse has been successfully calculated and 0 if src is singular.
RyoheiHagimoto	0:0e0631af0305	1767
RyoheiHagimoto	0:0e0631af0305	1768	In case of the DECOMP_SVD method, the function returns the inverse
RyoheiHagimoto	0:0e0631af0305	1769	condition number of src (the ratio of the smallest singular value to the
RyoheiHagimoto	0:0e0631af0305	1770	largest singular value) and 0 if src is singular. The SVD method
RyoheiHagimoto	0:0e0631af0305	1771	calculates a pseudo-inverse matrix if src is singular.
RyoheiHagimoto	0:0e0631af0305	1772
RyoheiHagimoto	0:0e0631af0305	1773	Similarly to DECOMP_LU, the method DECOMP_CHOLESKY works only with
RyoheiHagimoto	0:0e0631af0305	1774	non-singular square matrices that should also be symmetrical and
RyoheiHagimoto	0:0e0631af0305	1775	positively defined. In this case, the function stores the inverted
RyoheiHagimoto	0:0e0631af0305	1776	matrix in dst and returns non-zero. Otherwise, it returns 0.
RyoheiHagimoto	0:0e0631af0305	1777
RyoheiHagimoto	0:0e0631af0305	1778	@param src input floating-point M x N matrix.
RyoheiHagimoto	0:0e0631af0305	1779	@param dst output matrix of N x M size and the same type as src.
RyoheiHagimoto	0:0e0631af0305	1780	@param flags inversion method (cv::DecompTypes)
RyoheiHagimoto	0:0e0631af0305	1781	@sa solve, SVD
RyoheiHagimoto	0:0e0631af0305	1782	*/
RyoheiHagimoto	0:0e0631af0305	1783	CV_EXPORTS_W double invert(InputArray src, OutputArray dst, int flags = DECOMP_LU);
RyoheiHagimoto	0:0e0631af0305	1784
RyoheiHagimoto	0:0e0631af0305	1785	/** @brief Solves one or more linear systems or least-squares problems.
RyoheiHagimoto	0:0e0631af0305	1786
RyoheiHagimoto	0:0e0631af0305	1787	The function cv::solve solves a linear system or least-squares problem (the
RyoheiHagimoto	0:0e0631af0305	1788	latter is possible with SVD or QR methods, or by specifying the flag
RyoheiHagimoto	0:0e0631af0305	1789	DECOMP_NORMAL ):
RyoheiHagimoto	0:0e0631af0305	1790	\f[\texttt{dst} = \arg \min _X \\| \texttt{src1} \cdot \texttt{X} - \texttt{src2} \\|\f]
RyoheiHagimoto	0:0e0631af0305	1791
RyoheiHagimoto	0:0e0631af0305	1792	If DECOMP_LU or DECOMP_CHOLESKY method is used, the function returns 1
RyoheiHagimoto	0:0e0631af0305	1793	if src1 (or \f$\texttt{src1}^T\texttt{src1}\f$ ) is non-singular. Otherwise,
RyoheiHagimoto	0:0e0631af0305	1794	it returns 0. In the latter case, dst is not valid. Other methods find a
RyoheiHagimoto	0:0e0631af0305	1795	pseudo-solution in case of a singular left-hand side part.
RyoheiHagimoto	0:0e0631af0305	1796
RyoheiHagimoto	0:0e0631af0305	1797	@note If you want to find a unity-norm solution of an under-defined
RyoheiHagimoto	0:0e0631af0305	1798	singular system \f$\texttt{src1}\cdot\texttt{dst}=0\f$ , the function solve
RyoheiHagimoto	0:0e0631af0305	1799	will not do the work. Use SVD::solveZ instead.
RyoheiHagimoto	0:0e0631af0305	1800
RyoheiHagimoto	0:0e0631af0305	1801	@param src1 input matrix on the left-hand side of the system.
RyoheiHagimoto	0:0e0631af0305	1802	@param src2 input matrix on the right-hand side of the system.
RyoheiHagimoto	0:0e0631af0305	1803	@param dst output solution.
RyoheiHagimoto	0:0e0631af0305	1804	@param flags solution (matrix inversion) method (cv::DecompTypes)
RyoheiHagimoto	0:0e0631af0305	1805	@sa invert, SVD, eigen
RyoheiHagimoto	0:0e0631af0305	1806	*/
RyoheiHagimoto	0:0e0631af0305	1807	CV_EXPORTS_W bool solve(InputArray src1, InputArray src2,
RyoheiHagimoto	0:0e0631af0305	1808	OutputArray dst, int flags = DECOMP_LU);
RyoheiHagimoto	0:0e0631af0305	1809
RyoheiHagimoto	0:0e0631af0305	1810	/** @brief Sorts each row or each column of a matrix.
RyoheiHagimoto	0:0e0631af0305	1811
RyoheiHagimoto	0:0e0631af0305	1812	The function cv::sort sorts each matrix row or each matrix column in
RyoheiHagimoto	0:0e0631af0305	1813	ascending or descending order. So you should pass two operation flags to
RyoheiHagimoto	0:0e0631af0305	1814	get desired behaviour. If you want to sort matrix rows or columns
RyoheiHagimoto	0:0e0631af0305	1815	lexicographically, you can use STL std::sort generic function with the
RyoheiHagimoto	0:0e0631af0305	1816	proper comparison predicate.
RyoheiHagimoto	0:0e0631af0305	1817
RyoheiHagimoto	0:0e0631af0305	1818	@param src input single-channel array.
RyoheiHagimoto	0:0e0631af0305	1819	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	1820	@param flags operation flags, a combination of cv::SortFlags
RyoheiHagimoto	0:0e0631af0305	1821	@sa sortIdx, randShuffle
RyoheiHagimoto	0:0e0631af0305	1822	*/
RyoheiHagimoto	0:0e0631af0305	1823	CV_EXPORTS_W void sort(InputArray src, OutputArray dst, int flags);
RyoheiHagimoto	0:0e0631af0305	1824
RyoheiHagimoto	0:0e0631af0305	1825	/** @brief Sorts each row or each column of a matrix.
RyoheiHagimoto	0:0e0631af0305	1826
RyoheiHagimoto	0:0e0631af0305	1827	The function cv::sortIdx sorts each matrix row or each matrix column in the
RyoheiHagimoto	0:0e0631af0305	1828	ascending or descending order. So you should pass two operation flags to
RyoheiHagimoto	0:0e0631af0305	1829	get desired behaviour. Instead of reordering the elements themselves, it
RyoheiHagimoto	0:0e0631af0305	1830	stores the indices of sorted elements in the output array. For example:
RyoheiHagimoto	0:0e0631af0305	1831	@code
RyoheiHagimoto	0:0e0631af0305	1832	Mat A = Mat::eye(3,3,CV_32F), B;
RyoheiHagimoto	0:0e0631af0305	1833	sortIdx(A, B, SORT_EVERY_ROW + SORT_ASCENDING);
RyoheiHagimoto	0:0e0631af0305	1834	// B will probably contain
RyoheiHagimoto	0:0e0631af0305	1835	// (because of equal elements in A some permutations are possible):
RyoheiHagimoto	0:0e0631af0305	1836	// [[1, 2, 0], [0, 2, 1], [0, 1, 2]]
RyoheiHagimoto	0:0e0631af0305	1837	@endcode
RyoheiHagimoto	0:0e0631af0305	1838	@param src input single-channel array.
RyoheiHagimoto	0:0e0631af0305	1839	@param dst output integer array of the same size as src.
RyoheiHagimoto	0:0e0631af0305	1840	@param flags operation flags that could be a combination of cv::SortFlags
RyoheiHagimoto	0:0e0631af0305	1841	@sa sort, randShuffle
RyoheiHagimoto	0:0e0631af0305	1842	*/
RyoheiHagimoto	0:0e0631af0305	1843	CV_EXPORTS_W void sortIdx(InputArray src, OutputArray dst, int flags);
RyoheiHagimoto	0:0e0631af0305	1844
RyoheiHagimoto	0:0e0631af0305	1845	/** @brief Finds the real roots of a cubic equation.
RyoheiHagimoto	0:0e0631af0305	1846
RyoheiHagimoto	0:0e0631af0305	1847	The function solveCubic finds the real roots of a cubic equation:
RyoheiHagimoto	0:0e0631af0305	1848	- if coeffs is a 4-element vector:
RyoheiHagimoto	0:0e0631af0305	1849	\f[\texttt{coeffs} [0] x^3 + \texttt{coeffs} [1] x^2 + \texttt{coeffs} [2] x + \texttt{coeffs} [3] = 0\f]
RyoheiHagimoto	0:0e0631af0305	1850	- if coeffs is a 3-element vector:
RyoheiHagimoto	0:0e0631af0305	1851	\f[x^3 + \texttt{coeffs} [0] x^2 + \texttt{coeffs} [1] x + \texttt{coeffs} [2] = 0\f]
RyoheiHagimoto	0:0e0631af0305	1852
RyoheiHagimoto	0:0e0631af0305	1853	The roots are stored in the roots array.
RyoheiHagimoto	0:0e0631af0305	1854	@param coeffs equation coefficients, an array of 3 or 4 elements.
RyoheiHagimoto	0:0e0631af0305	1855	@param roots output array of real roots that has 1 or 3 elements.
RyoheiHagimoto	0:0e0631af0305	1856	*/
RyoheiHagimoto	0:0e0631af0305	1857	CV_EXPORTS_W int solveCubic(InputArray coeffs, OutputArray roots);
RyoheiHagimoto	0:0e0631af0305	1858
RyoheiHagimoto	0:0e0631af0305	1859	/** @brief Finds the real or complex roots of a polynomial equation.
RyoheiHagimoto	0:0e0631af0305	1860
RyoheiHagimoto	0:0e0631af0305	1861	The function cv::solvePoly finds real and complex roots of a polynomial equation:
RyoheiHagimoto	0:0e0631af0305	1862	\f[\texttt{coeffs} [n] x^{n} + \texttt{coeffs} [n-1] x^{n-1} + ... + \texttt{coeffs} [1] x + \texttt{coeffs} [0] = 0\f]
RyoheiHagimoto	0:0e0631af0305	1863	@param coeffs array of polynomial coefficients.
RyoheiHagimoto	0:0e0631af0305	1864	@param roots output (complex) array of roots.
RyoheiHagimoto	0:0e0631af0305	1865	@param maxIters maximum number of iterations the algorithm does.
RyoheiHagimoto	0:0e0631af0305	1866	*/
RyoheiHagimoto	0:0e0631af0305	1867	CV_EXPORTS_W double solvePoly(InputArray coeffs, OutputArray roots, int maxIters = 300);
RyoheiHagimoto	0:0e0631af0305	1868
RyoheiHagimoto	0:0e0631af0305	1869	/** @brief Calculates eigenvalues and eigenvectors of a symmetric matrix.
RyoheiHagimoto	0:0e0631af0305	1870
RyoheiHagimoto	0:0e0631af0305	1871	The function cv::eigen calculates just eigenvalues, or eigenvalues and eigenvectors of the symmetric
RyoheiHagimoto	0:0e0631af0305	1872	matrix src:
RyoheiHagimoto	0:0e0631af0305	1873	@code
RyoheiHagimoto	0:0e0631af0305	1874	srceigenvectors.row(i).t() = eigenvalues.at<srcType>(i)eigenvectors.row(i).t()
RyoheiHagimoto	0:0e0631af0305	1875	@endcode
RyoheiHagimoto	0:0e0631af0305	1876	@note in the new and the old interfaces different ordering of eigenvalues and eigenvectors
RyoheiHagimoto	0:0e0631af0305	1877	parameters is used.
RyoheiHagimoto	0:0e0631af0305	1878	@param src input matrix that must have CV_32FC1 or CV_64FC1 type, square size and be symmetrical
RyoheiHagimoto	0:0e0631af0305	1879	(src ^T^ == src).
RyoheiHagimoto	0:0e0631af0305	1880	@param eigenvalues output vector of eigenvalues of the same type as src; the eigenvalues are stored
RyoheiHagimoto	0:0e0631af0305	1881	in the descending order.
RyoheiHagimoto	0:0e0631af0305	1882	@param eigenvectors output matrix of eigenvectors; it has the same size and type as src; the
RyoheiHagimoto	0:0e0631af0305	1883	eigenvectors are stored as subsequent matrix rows, in the same order as the corresponding
RyoheiHagimoto	0:0e0631af0305	1884	eigenvalues.
RyoheiHagimoto	0:0e0631af0305	1885	@sa completeSymm , PCA
RyoheiHagimoto	0:0e0631af0305	1886	*/
RyoheiHagimoto	0:0e0631af0305	1887	CV_EXPORTS_W bool eigen(InputArray src, OutputArray eigenvalues,
RyoheiHagimoto	0:0e0631af0305	1888	OutputArray eigenvectors = noArray());
RyoheiHagimoto	0:0e0631af0305	1889
RyoheiHagimoto	0:0e0631af0305	1890	/** @brief Calculates the covariance matrix of a set of vectors.
RyoheiHagimoto	0:0e0631af0305	1891
RyoheiHagimoto	0:0e0631af0305	1892	The function cv::calcCovarMatrix calculates the covariance matrix and, optionally, the mean vector of
RyoheiHagimoto	0:0e0631af0305	1893	the set of input vectors.
RyoheiHagimoto	0:0e0631af0305	1894	@param samples samples stored as separate matrices
RyoheiHagimoto	0:0e0631af0305	1895	@param nsamples number of samples
RyoheiHagimoto	0:0e0631af0305	1896	@param covar output covariance matrix of the type ctype and square size.
RyoheiHagimoto	0:0e0631af0305	1897	@param mean input or output (depending on the flags) array as the average value of the input vectors.
RyoheiHagimoto	0:0e0631af0305	1898	@param flags operation flags as a combination of cv::CovarFlags
RyoheiHagimoto	0:0e0631af0305	1899	@param ctype type of the matrixl; it equals 'CV_64F' by default.
RyoheiHagimoto	0:0e0631af0305	1900	@sa PCA, mulTransposed, Mahalanobis
RyoheiHagimoto	0:0e0631af0305	1901	@todo InputArrayOfArrays
RyoheiHagimoto	0:0e0631af0305	1902	*/
RyoheiHagimoto	0:0e0631af0305	1903	CV_EXPORTS void calcCovarMatrix( const Mat* samples, int nsamples, Mat& covar, Mat& mean,
RyoheiHagimoto	0:0e0631af0305	1904	int flags, int ctype = CV_64F);
RyoheiHagimoto	0:0e0631af0305	1905
RyoheiHagimoto	0:0e0631af0305	1906	/** @overload
RyoheiHagimoto	0:0e0631af0305	1907	@note use cv::COVAR_ROWS or cv::COVAR_COLS flag
RyoheiHagimoto	0:0e0631af0305	1908	@param samples samples stored as rows/columns of a single matrix.
RyoheiHagimoto	0:0e0631af0305	1909	@param covar output covariance matrix of the type ctype and square size.
RyoheiHagimoto	0:0e0631af0305	1910	@param mean input or output (depending on the flags) array as the average value of the input vectors.
RyoheiHagimoto	0:0e0631af0305	1911	@param flags operation flags as a combination of cv::CovarFlags
RyoheiHagimoto	0:0e0631af0305	1912	@param ctype type of the matrixl; it equals 'CV_64F' by default.
RyoheiHagimoto	0:0e0631af0305	1913	*/
RyoheiHagimoto	0:0e0631af0305	1914	CV_EXPORTS_W void calcCovarMatrix( InputArray samples, OutputArray covar,
RyoheiHagimoto	0:0e0631af0305	1915	InputOutputArray mean, int flags, int ctype = CV_64F);
RyoheiHagimoto	0:0e0631af0305	1916
RyoheiHagimoto	0:0e0631af0305	1917	/** wrap PCA::operator() */
RyoheiHagimoto	0:0e0631af0305	1918	CV_EXPORTS_W void PCACompute(InputArray data, InputOutputArray mean,
RyoheiHagimoto	0:0e0631af0305	1919	OutputArray eigenvectors, int maxComponents = 0);
RyoheiHagimoto	0:0e0631af0305	1920
RyoheiHagimoto	0:0e0631af0305	1921	/** wrap PCA::operator() */
RyoheiHagimoto	0:0e0631af0305	1922	CV_EXPORTS_W void PCACompute(InputArray data, InputOutputArray mean,
RyoheiHagimoto	0:0e0631af0305	1923	OutputArray eigenvectors, double retainedVariance);
RyoheiHagimoto	0:0e0631af0305	1924
RyoheiHagimoto	0:0e0631af0305	1925	/** wrap PCA::project */
RyoheiHagimoto	0:0e0631af0305	1926	CV_EXPORTS_W void PCAProject(InputArray data, InputArray mean,
RyoheiHagimoto	0:0e0631af0305	1927	InputArray eigenvectors, OutputArray result);
RyoheiHagimoto	0:0e0631af0305	1928
RyoheiHagimoto	0:0e0631af0305	1929	/** wrap PCA::backProject */
RyoheiHagimoto	0:0e0631af0305	1930	CV_EXPORTS_W void PCABackProject(InputArray data, InputArray mean,
RyoheiHagimoto	0:0e0631af0305	1931	InputArray eigenvectors, OutputArray result);
RyoheiHagimoto	0:0e0631af0305	1932
RyoheiHagimoto	0:0e0631af0305	1933	/** wrap SVD::compute */
RyoheiHagimoto	0:0e0631af0305	1934	CV_EXPORTS_W void SVDecomp( InputArray src, OutputArray w, OutputArray u, OutputArray vt, int flags = 0 );
RyoheiHagimoto	0:0e0631af0305	1935
RyoheiHagimoto	0:0e0631af0305	1936	/** wrap SVD::backSubst */
RyoheiHagimoto	0:0e0631af0305	1937	CV_EXPORTS_W void SVBackSubst( InputArray w, InputArray u, InputArray vt,
RyoheiHagimoto	0:0e0631af0305	1938	InputArray rhs, OutputArray dst );
RyoheiHagimoto	0:0e0631af0305	1939
RyoheiHagimoto	0:0e0631af0305	1940	/** @brief Calculates the Mahalanobis distance between two vectors.
RyoheiHagimoto	0:0e0631af0305	1941
RyoheiHagimoto	0:0e0631af0305	1942	The function cv::Mahalanobis calculates and returns the weighted distance between two vectors:
RyoheiHagimoto	0:0e0631af0305	1943	\f[d( \texttt{vec1} , \texttt{vec2} )= \sqrt{\sum_{i,j}{\texttt{icovar(i,j)}\cdot(\texttt{vec1}(I)-\texttt{vec2}(I))\cdot(\texttt{vec1(j)}-\texttt{vec2(j)})} }\f]
RyoheiHagimoto	0:0e0631af0305	1944	The covariance matrix may be calculated using the cv::calcCovarMatrix function and then inverted using
RyoheiHagimoto	0:0e0631af0305	1945	the invert function (preferably using the cv::DECOMP_SVD method, as the most accurate).
RyoheiHagimoto	0:0e0631af0305	1946	@param v1 first 1D input vector.
RyoheiHagimoto	0:0e0631af0305	1947	@param v2 second 1D input vector.
RyoheiHagimoto	0:0e0631af0305	1948	@param icovar inverse covariance matrix.
RyoheiHagimoto	0:0e0631af0305	1949	*/
RyoheiHagimoto	0:0e0631af0305	1950	CV_EXPORTS_W double Mahalanobis(InputArray v1, InputArray v2, InputArray icovar);
RyoheiHagimoto	0:0e0631af0305	1951
RyoheiHagimoto	0:0e0631af0305	1952	/** @brief Performs a forward or inverse Discrete Fourier transform of a 1D or 2D floating-point array.
RyoheiHagimoto	0:0e0631af0305	1953
RyoheiHagimoto	0:0e0631af0305	1954	The function cv::dft performs one of the following:
RyoheiHagimoto	0:0e0631af0305	1955	- Forward the Fourier transform of a 1D vector of N elements:
RyoheiHagimoto	0:0e0631af0305	1956	\f[Y = F^{(N)} \cdot X,\f]
RyoheiHagimoto	0:0e0631af0305	1957	where \f$F^{(N)}_{jk}=\exp(-2\pi i j k/N)\f$ and \f$i=\sqrt{-1}\f$
RyoheiHagimoto	0:0e0631af0305	1958	- Inverse the Fourier transform of a 1D vector of N elements:
RyoheiHagimoto	0:0e0631af0305	1959	\f[\begin{array}{l} X'= \left (F^{(N)} \right )^{-1} \cdot Y = \left (F^{(N)} \right )^* \cdot y \\ X = (1/N) \cdot X, \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	1960	where \f$F^*=\left(\textrm{Re}(F^{(N)})-\textrm{Im}(F^{(N)})\right)^T\f$
RyoheiHagimoto	0:0e0631af0305	1961	- Forward the 2D Fourier transform of a M x N matrix:
RyoheiHagimoto	0:0e0631af0305	1962	\f[Y = F^{(M)} \cdot X \cdot F^{(N)}\f]
RyoheiHagimoto	0:0e0631af0305	1963	- Inverse the 2D Fourier transform of a M x N matrix:
RyoheiHagimoto	0:0e0631af0305	1964	\f[\begin{array}{l} X'= \left (F^{(M)} \right )^* \cdot Y \cdot \left (F^{(N)} \right )^* \\ X = \frac{1}{M \cdot N} \cdot X' \end{array}\f]
RyoheiHagimoto	0:0e0631af0305	1965
RyoheiHagimoto	0:0e0631af0305	1966	In case of real (single-channel) data, the output spectrum of the forward Fourier transform or input
RyoheiHagimoto	0:0e0631af0305	1967	spectrum of the inverse Fourier transform can be represented in a packed format called CCS
RyoheiHagimoto	0:0e0631af0305	1968	(complex-conjugate-symmetrical). It was borrowed from IPL (Intel\* Image Processing Library). Here
RyoheiHagimoto	0:0e0631af0305	1969	is how 2D CCS spectrum looks:
RyoheiHagimoto	0:0e0631af0305	1970	\f[\begin{bmatrix} Re Y_{0,0} & Re Y_{0,1} & Im Y_{0,1} & Re Y_{0,2} & Im Y_{0,2} & \cdots & Re Y_{0,N/2-1} & Im Y_{0,N/2-1} & Re Y_{0,N/2} \\ Re Y_{1,0} & Re Y_{1,1} & Im Y_{1,1} & Re Y_{1,2} & Im Y_{1,2} & \cdots & Re Y_{1,N/2-1} & Im Y_{1,N/2-1} & Re Y_{1,N/2} \\ Im Y_{1,0} & Re Y_{2,1} & Im Y_{2,1} & Re Y_{2,2} & Im Y_{2,2} & \cdots & Re Y_{2,N/2-1} & Im Y_{2,N/2-1} & Im Y_{1,N/2} \\ \hdotsfor{9} \\ Re Y_{M/2-1,0} & Re Y_{M-3,1} & Im Y_{M-3,1} & \hdotsfor{3} & Re Y_{M-3,N/2-1} & Im Y_{M-3,N/2-1}& Re Y_{M/2-1,N/2} \\ Im Y_{M/2-1,0} & Re Y_{M-2,1} & Im Y_{M-2,1} & \hdotsfor{3} & Re Y_{M-2,N/2-1} & Im Y_{M-2,N/2-1}& Im Y_{M/2-1,N/2} \\ Re Y_{M/2,0} & Re Y_{M-1,1} & Im Y_{M-1,1} & \hdotsfor{3} & Re Y_{M-1,N/2-1} & Im Y_{M-1,N/2-1}& Re Y_{M/2,N/2} \end{bmatrix}\f]
RyoheiHagimoto	0:0e0631af0305	1971
RyoheiHagimoto	0:0e0631af0305	1972	In case of 1D transform of a real vector, the output looks like the first row of the matrix above.
RyoheiHagimoto	0:0e0631af0305	1973
RyoheiHagimoto	0:0e0631af0305	1974	So, the function chooses an operation mode depending on the flags and size of the input array:
RyoheiHagimoto	0:0e0631af0305	1975	- If DFT_ROWS is set or the input array has a single row or single column, the function
RyoheiHagimoto	0:0e0631af0305	1976	performs a 1D forward or inverse transform of each row of a matrix when DFT_ROWS is set.
RyoheiHagimoto	0:0e0631af0305	1977	Otherwise, it performs a 2D transform.
RyoheiHagimoto	0:0e0631af0305	1978	- If the input array is real and DFT_INVERSE is not set, the function performs a forward 1D or
RyoheiHagimoto	0:0e0631af0305	1979	2D transform:
RyoheiHagimoto	0:0e0631af0305	1980	- When DFT_COMPLEX_OUTPUT is set, the output is a complex matrix of the same size as
RyoheiHagimoto	0:0e0631af0305	1981	input.
RyoheiHagimoto	0:0e0631af0305	1982	- When DFT_COMPLEX_OUTPUT is not set, the output is a real matrix of the same size as
RyoheiHagimoto	0:0e0631af0305	1983	input. In case of 2D transform, it uses the packed format as shown above. In case of a
RyoheiHagimoto	0:0e0631af0305	1984	single 1D transform, it looks like the first row of the matrix above. In case of
RyoheiHagimoto	0:0e0631af0305	1985	multiple 1D transforms (when using the DFT_ROWS flag), each row of the output matrix
RyoheiHagimoto	0:0e0631af0305	1986	looks like the first row of the matrix above.
RyoheiHagimoto	0:0e0631af0305	1987	- If the input array is complex and either DFT_INVERSE or DFT_REAL_OUTPUT are not set, the
RyoheiHagimoto	0:0e0631af0305	1988	output is a complex array of the same size as input. The function performs a forward or
RyoheiHagimoto	0:0e0631af0305	1989	inverse 1D or 2D transform of the whole input array or each row of the input array
RyoheiHagimoto	0:0e0631af0305	1990	independently, depending on the flags DFT_INVERSE and DFT_ROWS.
RyoheiHagimoto	0:0e0631af0305	1991	- When DFT_INVERSE is set and the input array is real, or it is complex but DFT_REAL_OUTPUT
RyoheiHagimoto	0:0e0631af0305	1992	is set, the output is a real array of the same size as input. The function performs a 1D or 2D
RyoheiHagimoto	0:0e0631af0305	1993	inverse transformation of the whole input array or each individual row, depending on the flags
RyoheiHagimoto	0:0e0631af0305	1994	DFT_INVERSE and DFT_ROWS.
RyoheiHagimoto	0:0e0631af0305	1995
RyoheiHagimoto	0:0e0631af0305	1996	If DFT_SCALE is set, the scaling is done after the transformation.
RyoheiHagimoto	0:0e0631af0305	1997
RyoheiHagimoto	0:0e0631af0305	1998	Unlike dct , the function supports arrays of arbitrary size. But only those arrays are processed
RyoheiHagimoto	0:0e0631af0305	1999	efficiently, whose sizes can be factorized in a product of small prime numbers (2, 3, and 5 in the
RyoheiHagimoto	0:0e0631af0305	2000	current implementation). Such an efficient DFT size can be calculated using the getOptimalDFTSize
RyoheiHagimoto	0:0e0631af0305	2001	method.
RyoheiHagimoto	0:0e0631af0305	2002
RyoheiHagimoto	0:0e0631af0305	2003	The sample below illustrates how to calculate a DFT-based convolution of two 2D real arrays:
RyoheiHagimoto	0:0e0631af0305	2004	@code
RyoheiHagimoto	0:0e0631af0305	2005	void convolveDFT(InputArray A, InputArray B, OutputArray C)
RyoheiHagimoto	0:0e0631af0305	2006	{
RyoheiHagimoto	0:0e0631af0305	2007	// reallocate the output array if needed
RyoheiHagimoto	0:0e0631af0305	2008	C.create(abs(A.rows - B.rows)+1, abs(A.cols - B.cols)+1, A.type());
RyoheiHagimoto	0:0e0631af0305	2009	Size dftSize;
RyoheiHagimoto	0:0e0631af0305	2010	// calculate the size of DFT transform
RyoheiHagimoto	0:0e0631af0305	2011	dftSize.width = getOptimalDFTSize(A.cols + B.cols - 1);
RyoheiHagimoto	0:0e0631af0305	2012	dftSize.height = getOptimalDFTSize(A.rows + B.rows - 1);
RyoheiHagimoto	0:0e0631af0305	2013
RyoheiHagimoto	0:0e0631af0305	2014	// allocate temporary buffers and initialize them with 0's
RyoheiHagimoto	0:0e0631af0305	2015	Mat tempA(dftSize, A.type(), Scalar::all(0));
RyoheiHagimoto	0:0e0631af0305	2016	Mat tempB(dftSize, B.type(), Scalar::all(0));
RyoheiHagimoto	0:0e0631af0305	2017
RyoheiHagimoto	0:0e0631af0305	2018	// copy A and B to the top-left corners of tempA and tempB, respectively
RyoheiHagimoto	0:0e0631af0305	2019	Mat roiA(tempA, Rect(0,0,A.cols,A.rows));
RyoheiHagimoto	0:0e0631af0305	2020	A.copyTo(roiA);
RyoheiHagimoto	0:0e0631af0305	2021	Mat roiB(tempB, Rect(0,0,B.cols,B.rows));
RyoheiHagimoto	0:0e0631af0305	2022	B.copyTo(roiB);
RyoheiHagimoto	0:0e0631af0305	2023
RyoheiHagimoto	0:0e0631af0305	2024	// now transform the padded A & B in-place;
RyoheiHagimoto	0:0e0631af0305	2025	// use "nonzeroRows" hint for faster processing
RyoheiHagimoto	0:0e0631af0305	2026	dft(tempA, tempA, 0, A.rows);
RyoheiHagimoto	0:0e0631af0305	2027	dft(tempB, tempB, 0, B.rows);
RyoheiHagimoto	0:0e0631af0305	2028
RyoheiHagimoto	0:0e0631af0305	2029	// multiply the spectrums;
RyoheiHagimoto	0:0e0631af0305	2030	// the function handles packed spectrum representations well
RyoheiHagimoto	0:0e0631af0305	2031	mulSpectrums(tempA, tempB, tempA);
RyoheiHagimoto	0:0e0631af0305	2032
RyoheiHagimoto	0:0e0631af0305	2033	// transform the product back from the frequency domain.
RyoheiHagimoto	0:0e0631af0305	2034	// Even though all the result rows will be non-zero,
RyoheiHagimoto	0:0e0631af0305	2035	// you need only the first C.rows of them, and thus you
RyoheiHagimoto	0:0e0631af0305	2036	// pass nonzeroRows == C.rows
RyoheiHagimoto	0:0e0631af0305	2037	dft(tempA, tempA, DFT_INVERSE + DFT_SCALE, C.rows);
RyoheiHagimoto	0:0e0631af0305	2038
RyoheiHagimoto	0:0e0631af0305	2039	// now copy the result back to C.
RyoheiHagimoto	0:0e0631af0305	2040	tempA(Rect(0, 0, C.cols, C.rows)).copyTo(C);
RyoheiHagimoto	0:0e0631af0305	2041
RyoheiHagimoto	0:0e0631af0305	2042	// all the temporary buffers will be deallocated automatically
RyoheiHagimoto	0:0e0631af0305	2043	}
RyoheiHagimoto	0:0e0631af0305	2044	@endcode
RyoheiHagimoto	0:0e0631af0305	2045	To optimize this sample, consider the following approaches:
RyoheiHagimoto	0:0e0631af0305	2046	- Since nonzeroRows != 0 is passed to the forward transform calls and since A and B are copied to
RyoheiHagimoto	0:0e0631af0305	2047	the top-left corners of tempA and tempB, respectively, it is not necessary to clear the whole
RyoheiHagimoto	0:0e0631af0305	2048	tempA and tempB. It is only necessary to clear the tempA.cols - A.cols ( tempB.cols - B.cols)
RyoheiHagimoto	0:0e0631af0305	2049	rightmost columns of the matrices.
RyoheiHagimoto	0:0e0631af0305	2050	- This DFT-based convolution does not have to be applied to the whole big arrays, especially if B
RyoheiHagimoto	0:0e0631af0305	2051	is significantly smaller than A or vice versa. Instead, you can calculate convolution by parts.
RyoheiHagimoto	0:0e0631af0305	2052	To do this, you need to split the output array C into multiple tiles. For each tile, estimate
RyoheiHagimoto	0:0e0631af0305	2053	which parts of A and B are required to calculate convolution in this tile. If the tiles in C are
RyoheiHagimoto	0:0e0631af0305	2054	too small, the speed will decrease a lot because of repeated work. In the ultimate case, when
RyoheiHagimoto	0:0e0631af0305	2055	each tile in C is a single pixel, the algorithm becomes equivalent to the naive convolution
RyoheiHagimoto	0:0e0631af0305	2056	algorithm. If the tiles are too big, the temporary arrays tempA and tempB become too big and
RyoheiHagimoto	0:0e0631af0305	2057	there is also a slowdown because of bad cache locality. So, there is an optimal tile size
RyoheiHagimoto	0:0e0631af0305	2058	somewhere in the middle.
RyoheiHagimoto	0:0e0631af0305	2059	- If different tiles in C can be calculated in parallel and, thus, the convolution is done by
RyoheiHagimoto	0:0e0631af0305	2060	parts, the loop can be threaded.
RyoheiHagimoto	0:0e0631af0305	2061
RyoheiHagimoto	0:0e0631af0305	2062	All of the above improvements have been implemented in matchTemplate and filter2D . Therefore, by
RyoheiHagimoto	0:0e0631af0305	2063	using them, you can get the performance even better than with the above theoretically optimal
RyoheiHagimoto	0:0e0631af0305	2064	implementation. Though, those two functions actually calculate cross-correlation, not convolution,
RyoheiHagimoto	0:0e0631af0305	2065	so you need to "flip" the second convolution operand B vertically and horizontally using flip .
RyoheiHagimoto	0:0e0631af0305	2066	@note
RyoheiHagimoto	0:0e0631af0305	2067	- An example using the discrete fourier transform can be found at
RyoheiHagimoto	0:0e0631af0305	2068	opencv_source_code/samples/cpp/dft.cpp
RyoheiHagimoto	0:0e0631af0305	2069	- (Python) An example using the dft functionality to perform Wiener deconvolution can be found
RyoheiHagimoto	0:0e0631af0305	2070	at opencv_source/samples/python/deconvolution.py
RyoheiHagimoto	0:0e0631af0305	2071	- (Python) An example rearranging the quadrants of a Fourier image can be found at
RyoheiHagimoto	0:0e0631af0305	2072	opencv_source/samples/python/dft.py
RyoheiHagimoto	0:0e0631af0305	2073	@param src input array that could be real or complex.
RyoheiHagimoto	0:0e0631af0305	2074	@param dst output array whose size and type depends on the flags .
RyoheiHagimoto	0:0e0631af0305	2075	@param flags transformation flags, representing a combination of the cv::DftFlags
RyoheiHagimoto	0:0e0631af0305	2076	@param nonzeroRows when the parameter is not zero, the function assumes that only the first
RyoheiHagimoto	0:0e0631af0305	2077	nonzeroRows rows of the input array (DFT_INVERSE is not set) or only the first nonzeroRows of the
RyoheiHagimoto	0:0e0631af0305	2078	output array (DFT_INVERSE is set) contain non-zeros, thus, the function can handle the rest of the
RyoheiHagimoto	0:0e0631af0305	2079	rows more efficiently and save some time; this technique is very useful for calculating array
RyoheiHagimoto	0:0e0631af0305	2080	cross-correlation or convolution using DFT.
RyoheiHagimoto	0:0e0631af0305	2081	@sa dct , getOptimalDFTSize , mulSpectrums, filter2D , matchTemplate , flip , cartToPolar ,
RyoheiHagimoto	0:0e0631af0305	2082	magnitude , phase
RyoheiHagimoto	0:0e0631af0305	2083	*/
RyoheiHagimoto	0:0e0631af0305	2084	CV_EXPORTS_W void dft(InputArray src, OutputArray dst, int flags = 0, int nonzeroRows = 0);
RyoheiHagimoto	0:0e0631af0305	2085
RyoheiHagimoto	0:0e0631af0305	2086	/** @brief Calculates the inverse Discrete Fourier Transform of a 1D or 2D array.
RyoheiHagimoto	0:0e0631af0305	2087
RyoheiHagimoto	0:0e0631af0305	2088	idft(src, dst, flags) is equivalent to dft(src, dst, flags \| DFT_INVERSE) .
RyoheiHagimoto	0:0e0631af0305	2089	@note None of dft and idft scales the result by default. So, you should pass DFT_SCALE to one of
RyoheiHagimoto	0:0e0631af0305	2090	dft or idft explicitly to make these transforms mutually inverse.
RyoheiHagimoto	0:0e0631af0305	2091	@sa dft, dct, idct, mulSpectrums, getOptimalDFTSize
RyoheiHagimoto	0:0e0631af0305	2092	@param src input floating-point real or complex array.
RyoheiHagimoto	0:0e0631af0305	2093	@param dst output array whose size and type depend on the flags.
RyoheiHagimoto	0:0e0631af0305	2094	@param flags operation flags (see dft and cv::DftFlags).
RyoheiHagimoto	0:0e0631af0305	2095	@param nonzeroRows number of dst rows to process; the rest of the rows have undefined content (see
RyoheiHagimoto	0:0e0631af0305	2096	the convolution sample in dft description.
RyoheiHagimoto	0:0e0631af0305	2097	*/
RyoheiHagimoto	0:0e0631af0305	2098	CV_EXPORTS_W void idft(InputArray src, OutputArray dst, int flags = 0, int nonzeroRows = 0);
RyoheiHagimoto	0:0e0631af0305	2099
RyoheiHagimoto	0:0e0631af0305	2100	/** @brief Performs a forward or inverse discrete Cosine transform of 1D or 2D array.
RyoheiHagimoto	0:0e0631af0305	2101
RyoheiHagimoto	0:0e0631af0305	2102	The function cv::dct performs a forward or inverse discrete Cosine transform (DCT) of a 1D or 2D
RyoheiHagimoto	0:0e0631af0305	2103	floating-point array:
RyoheiHagimoto	0:0e0631af0305	2104	- Forward Cosine transform of a 1D vector of N elements:
RyoheiHagimoto	0:0e0631af0305	2105	\f[Y = C^{(N)} \cdot X\f]
RyoheiHagimoto	0:0e0631af0305	2106	where
RyoheiHagimoto	0:0e0631af0305	2107	\f[C^{(N)}_{jk}= \sqrt{\alpha_j/N} \cos \left ( \frac{\pi(2k+1)j}{2N} \right )\f]
RyoheiHagimoto	0:0e0631af0305	2108	and
RyoheiHagimoto	0:0e0631af0305	2109	\f$\alpha_0=1\f$, \f$\alpha_j=2\f$ for j \> 0.
RyoheiHagimoto	0:0e0631af0305	2110	- Inverse Cosine transform of a 1D vector of N elements:
RyoheiHagimoto	0:0e0631af0305	2111	\f[X = \left (C^{(N)} \right )^{-1} \cdot Y = \left (C^{(N)} \right )^T \cdot Y\f]
RyoheiHagimoto	0:0e0631af0305	2112	(since \f$C^{(N)}\f$ is an orthogonal matrix, \f$C^{(N)} \cdot \left(C^{(N)}\right)^T = I\f$ )
RyoheiHagimoto	0:0e0631af0305	2113	- Forward 2D Cosine transform of M x N matrix:
RyoheiHagimoto	0:0e0631af0305	2114	\f[Y = C^{(N)} \cdot X \cdot \left (C^{(N)} \right )^T\f]
RyoheiHagimoto	0:0e0631af0305	2115	- Inverse 2D Cosine transform of M x N matrix:
RyoheiHagimoto	0:0e0631af0305	2116	\f[X = \left (C^{(N)} \right )^T \cdot X \cdot C^{(N)}\f]
RyoheiHagimoto	0:0e0631af0305	2117
RyoheiHagimoto	0:0e0631af0305	2118	The function chooses the mode of operation by looking at the flags and size of the input array:
RyoheiHagimoto	0:0e0631af0305	2119	- If (flags & DCT_INVERSE) == 0 , the function does a forward 1D or 2D transform. Otherwise, it
RyoheiHagimoto	0:0e0631af0305	2120	is an inverse 1D or 2D transform.
RyoheiHagimoto	0:0e0631af0305	2121	- If (flags & DCT_ROWS) != 0 , the function performs a 1D transform of each row.
RyoheiHagimoto	0:0e0631af0305	2122	- If the array is a single column or a single row, the function performs a 1D transform.
RyoheiHagimoto	0:0e0631af0305	2123	- If none of the above is true, the function performs a 2D transform.
RyoheiHagimoto	0:0e0631af0305	2124
RyoheiHagimoto	0:0e0631af0305	2125	@note Currently dct supports even-size arrays (2, 4, 6 ...). For data analysis and approximation, you
RyoheiHagimoto	0:0e0631af0305	2126	can pad the array when necessary.
RyoheiHagimoto	0:0e0631af0305	2127	Also, the function performance depends very much, and not monotonically, on the array size (see
RyoheiHagimoto	0:0e0631af0305	2128	getOptimalDFTSize ). In the current implementation DCT of a vector of size N is calculated via DFT
RyoheiHagimoto	0:0e0631af0305	2129	of a vector of size N/2 . Thus, the optimal DCT size N1 \>= N can be calculated as:
RyoheiHagimoto	0:0e0631af0305	2130	@code
RyoheiHagimoto	0:0e0631af0305	2131	size_t getOptimalDCTSize(size_t N) { return 2*getOptimalDFTSize((N+1)/2); }
RyoheiHagimoto	0:0e0631af0305	2132	N1 = getOptimalDCTSize(N);
RyoheiHagimoto	0:0e0631af0305	2133	@endcode
RyoheiHagimoto	0:0e0631af0305	2134	@param src input floating-point array.
RyoheiHagimoto	0:0e0631af0305	2135	@param dst output array of the same size and type as src .
RyoheiHagimoto	0:0e0631af0305	2136	@param flags transformation flags as a combination of cv::DftFlags (DCT_*)
RyoheiHagimoto	0:0e0631af0305	2137	@sa dft , getOptimalDFTSize , idct
RyoheiHagimoto	0:0e0631af0305	2138	*/
RyoheiHagimoto	0:0e0631af0305	2139	CV_EXPORTS_W void dct(InputArray src, OutputArray dst, int flags = 0);
RyoheiHagimoto	0:0e0631af0305	2140
RyoheiHagimoto	0:0e0631af0305	2141	/** @brief Calculates the inverse Discrete Cosine Transform of a 1D or 2D array.
RyoheiHagimoto	0:0e0631af0305	2142
RyoheiHagimoto	0:0e0631af0305	2143	idct(src, dst, flags) is equivalent to dct(src, dst, flags \| DCT_INVERSE).
RyoheiHagimoto	0:0e0631af0305	2144	@param src input floating-point single-channel array.
RyoheiHagimoto	0:0e0631af0305	2145	@param dst output array of the same size and type as src.
RyoheiHagimoto	0:0e0631af0305	2146	@param flags operation flags.
RyoheiHagimoto	0:0e0631af0305	2147	@sa dct, dft, idft, getOptimalDFTSize
RyoheiHagimoto	0:0e0631af0305	2148	*/
RyoheiHagimoto	0:0e0631af0305	2149	CV_EXPORTS_W void idct(InputArray src, OutputArray dst, int flags = 0);
RyoheiHagimoto	0:0e0631af0305	2150
RyoheiHagimoto	0:0e0631af0305	2151	/** @brief Performs the per-element multiplication of two Fourier spectrums.
RyoheiHagimoto	0:0e0631af0305	2152
RyoheiHagimoto	0:0e0631af0305	2153	The function cv::mulSpectrums performs the per-element multiplication of the two CCS-packed or complex
RyoheiHagimoto	0:0e0631af0305	2154	matrices that are results of a real or complex Fourier transform.
RyoheiHagimoto	0:0e0631af0305	2155
RyoheiHagimoto	0:0e0631af0305	2156	The function, together with dft and idft , may be used to calculate convolution (pass conjB=false )
RyoheiHagimoto	0:0e0631af0305	2157	or correlation (pass conjB=true ) of two arrays rapidly. When the arrays are complex, they are
RyoheiHagimoto	0:0e0631af0305	2158	simply multiplied (per element) with an optional conjugation of the second-array elements. When the
RyoheiHagimoto	0:0e0631af0305	2159	arrays are real, they are assumed to be CCS-packed (see dft for details).
RyoheiHagimoto	0:0e0631af0305	2160	@param a first input array.
RyoheiHagimoto	0:0e0631af0305	2161	@param b second input array of the same size and type as src1 .
RyoheiHagimoto	0:0e0631af0305	2162	@param c output array of the same size and type as src1 .
RyoheiHagimoto	0:0e0631af0305	2163	@param flags operation flags; currently, the only supported flag is cv::DFT_ROWS, which indicates that
RyoheiHagimoto	0:0e0631af0305	2164	each row of src1 and src2 is an independent 1D Fourier spectrum. If you do not want to use this flag, then simply add a `0` as value.
RyoheiHagimoto	0:0e0631af0305	2165	@param conjB optional flag that conjugates the second input array before the multiplication (true)
RyoheiHagimoto	0:0e0631af0305	2166	or not (false).
RyoheiHagimoto	0:0e0631af0305	2167	*/
RyoheiHagimoto	0:0e0631af0305	2168	CV_EXPORTS_W void mulSpectrums(InputArray a, InputArray b, OutputArray c,
RyoheiHagimoto	0:0e0631af0305	2169	int flags, bool conjB = false);
RyoheiHagimoto	0:0e0631af0305	2170
RyoheiHagimoto	0:0e0631af0305	2171	/** @brief Returns the optimal DFT size for a given vector size.
RyoheiHagimoto	0:0e0631af0305	2172
RyoheiHagimoto	0:0e0631af0305	2173	DFT performance is not a monotonic function of a vector size. Therefore, when you calculate
RyoheiHagimoto	0:0e0631af0305	2174	convolution of two arrays or perform the spectral analysis of an array, it usually makes sense to
RyoheiHagimoto	0:0e0631af0305	2175	pad the input data with zeros to get a bit larger array that can be transformed much faster than the
RyoheiHagimoto	0:0e0631af0305	2176	original one. Arrays whose size is a power-of-two (2, 4, 8, 16, 32, ...) are the fastest to process.
RyoheiHagimoto	0:0e0631af0305	2177	Though, the arrays whose size is a product of 2's, 3's, and 5's (for example, 300 = 5\5\3\2\2)
RyoheiHagimoto	0:0e0631af0305	2178	are also processed quite efficiently.
RyoheiHagimoto	0:0e0631af0305	2179
RyoheiHagimoto	0:0e0631af0305	2180	The function cv::getOptimalDFTSize returns the minimum number N that is greater than or equal to vecsize
RyoheiHagimoto	0:0e0631af0305	2181	so that the DFT of a vector of size N can be processed efficiently. In the current implementation N
RyoheiHagimoto	0:0e0631af0305	2182	= 2 ^p^ \* 3 ^q^ \* 5 ^r^ for some integer p, q, r.
RyoheiHagimoto	0:0e0631af0305	2183
RyoheiHagimoto	0:0e0631af0305	2184	The function returns a negative number if vecsize is too large (very close to INT_MAX ).
RyoheiHagimoto	0:0e0631af0305	2185
RyoheiHagimoto	0:0e0631af0305	2186	While the function cannot be used directly to estimate the optimal vector size for DCT transform
RyoheiHagimoto	0:0e0631af0305	2187	(since the current DCT implementation supports only even-size vectors), it can be easily processed
RyoheiHagimoto	0:0e0631af0305	2188	as getOptimalDFTSize((vecsize+1)/2)\*2.
RyoheiHagimoto	0:0e0631af0305	2189	@param vecsize vector size.
RyoheiHagimoto	0:0e0631af0305	2190	@sa dft , dct , idft , idct , mulSpectrums
RyoheiHagimoto	0:0e0631af0305	2191	*/
RyoheiHagimoto	0:0e0631af0305	2192	CV_EXPORTS_W int getOptimalDFTSize(int vecsize);
RyoheiHagimoto	0:0e0631af0305	2193
RyoheiHagimoto	0:0e0631af0305	2194	/** @brief Returns the default random number generator.
RyoheiHagimoto	0:0e0631af0305	2195
RyoheiHagimoto	0:0e0631af0305	2196	The function cv::theRNG returns the default random number generator. For each thread, there is a
RyoheiHagimoto	0:0e0631af0305	2197	separate random number generator, so you can use the function safely in multi-thread environments.
RyoheiHagimoto	0:0e0631af0305	2198	If you just need to get a single random number using this generator or initialize an array, you can
RyoheiHagimoto	0:0e0631af0305	2199	use randu or randn instead. But if you are going to generate many random numbers inside a loop, it
RyoheiHagimoto	0:0e0631af0305	2200	is much faster to use this function to retrieve the generator and then use RNG::operator _Tp() .
RyoheiHagimoto	0:0e0631af0305	2201	@sa RNG, randu, randn
RyoheiHagimoto	0:0e0631af0305	2202	*/
RyoheiHagimoto	0:0e0631af0305	2203	CV_EXPORTS RNG& theRNG();
RyoheiHagimoto	0:0e0631af0305	2204
RyoheiHagimoto	0:0e0631af0305	2205	/** @brief Sets state of default random number generator.
RyoheiHagimoto	0:0e0631af0305	2206
RyoheiHagimoto	0:0e0631af0305	2207	The function cv::setRNGSeed sets state of default random number generator to custom value.
RyoheiHagimoto	0:0e0631af0305	2208	@param seed new state for default random number generator
RyoheiHagimoto	0:0e0631af0305	2209	@sa RNG, randu, randn
RyoheiHagimoto	0:0e0631af0305	2210	*/
RyoheiHagimoto	0:0e0631af0305	2211	CV_EXPORTS_W void setRNGSeed(int seed);
RyoheiHagimoto	0:0e0631af0305	2212
RyoheiHagimoto	0:0e0631af0305	2213	/** @brief Generates a single uniformly-distributed random number or an array of random numbers.
RyoheiHagimoto	0:0e0631af0305	2214
RyoheiHagimoto	0:0e0631af0305	2215	Non-template variant of the function fills the matrix dst with uniformly-distributed
RyoheiHagimoto	0:0e0631af0305	2216	random numbers from the specified range:
RyoheiHagimoto	0:0e0631af0305	2217	\f[\texttt{low} _c \leq \texttt{dst} (I)_c < \texttt{high} _c\f]
RyoheiHagimoto	0:0e0631af0305	2218	@param dst output array of random numbers; the array must be pre-allocated.
RyoheiHagimoto	0:0e0631af0305	2219	@param low inclusive lower boundary of the generated random numbers.
RyoheiHagimoto	0:0e0631af0305	2220	@param high exclusive upper boundary of the generated random numbers.
RyoheiHagimoto	0:0e0631af0305	2221	@sa RNG, randn, theRNG
RyoheiHagimoto	0:0e0631af0305	2222	*/
RyoheiHagimoto	0:0e0631af0305	2223	CV_EXPORTS_W void randu(InputOutputArray dst, InputArray low, InputArray high);
RyoheiHagimoto	0:0e0631af0305	2224
RyoheiHagimoto	0:0e0631af0305	2225	/** @brief Fills the array with normally distributed random numbers.
RyoheiHagimoto	0:0e0631af0305	2226
RyoheiHagimoto	0:0e0631af0305	2227	The function cv::randn fills the matrix dst with normally distributed random numbers with the specified
RyoheiHagimoto	0:0e0631af0305	2228	mean vector and the standard deviation matrix. The generated random numbers are clipped to fit the
RyoheiHagimoto	0:0e0631af0305	2229	value range of the output array data type.
RyoheiHagimoto	0:0e0631af0305	2230	@param dst output array of random numbers; the array must be pre-allocated and have 1 to 4 channels.
RyoheiHagimoto	0:0e0631af0305	2231	@param mean mean value (expectation) of the generated random numbers.
RyoheiHagimoto	0:0e0631af0305	2232	@param stddev standard deviation of the generated random numbers; it can be either a vector (in
RyoheiHagimoto	0:0e0631af0305	2233	which case a diagonal standard deviation matrix is assumed) or a square matrix.
RyoheiHagimoto	0:0e0631af0305	2234	@sa RNG, randu
RyoheiHagimoto	0:0e0631af0305	2235	*/
RyoheiHagimoto	0:0e0631af0305	2236	CV_EXPORTS_W void randn(InputOutputArray dst, InputArray mean, InputArray stddev);
RyoheiHagimoto	0:0e0631af0305	2237
RyoheiHagimoto	0:0e0631af0305	2238	/** @brief Shuffles the array elements randomly.
RyoheiHagimoto	0:0e0631af0305	2239
RyoheiHagimoto	0:0e0631af0305	2240	The function cv::randShuffle shuffles the specified 1D array by randomly choosing pairs of elements and
RyoheiHagimoto	0:0e0631af0305	2241	swapping them. The number of such swap operations will be dst.rows\dst.cols\iterFactor .
RyoheiHagimoto	0:0e0631af0305	2242	@param dst input/output numerical 1D array.
RyoheiHagimoto	0:0e0631af0305	2243	@param iterFactor scale factor that determines the number of random swap operations (see the details
RyoheiHagimoto	0:0e0631af0305	2244	below).
RyoheiHagimoto	0:0e0631af0305	2245	@param rng optional random number generator used for shuffling; if it is zero, theRNG () is used
RyoheiHagimoto	0:0e0631af0305	2246	instead.
RyoheiHagimoto	0:0e0631af0305	2247	@sa RNG, sort
RyoheiHagimoto	0:0e0631af0305	2248	*/
RyoheiHagimoto	0:0e0631af0305	2249	CV_EXPORTS_W void randShuffle(InputOutputArray dst, double iterFactor = 1., RNG* rng = 0);
RyoheiHagimoto	0:0e0631af0305	2250
RyoheiHagimoto	0:0e0631af0305	2251	/** @brief Principal Component Analysis
RyoheiHagimoto	0:0e0631af0305	2252
RyoheiHagimoto	0:0e0631af0305	2253	The class is used to calculate a special basis for a set of vectors. The
RyoheiHagimoto	0:0e0631af0305	2254	basis will consist of eigenvectors of the covariance matrix calculated
RyoheiHagimoto	0:0e0631af0305	2255	from the input set of vectors. The class %PCA can also transform
RyoheiHagimoto	0:0e0631af0305	2256	vectors to/from the new coordinate space defined by the basis. Usually,
RyoheiHagimoto	0:0e0631af0305	2257	in this new coordinate system, each vector from the original set (and
RyoheiHagimoto	0:0e0631af0305	2258	any linear combination of such vectors) can be quite accurately
RyoheiHagimoto	0:0e0631af0305	2259	approximated by taking its first few components, corresponding to the
RyoheiHagimoto	0:0e0631af0305	2260	eigenvectors of the largest eigenvalues of the covariance matrix.
RyoheiHagimoto	0:0e0631af0305	2261	Geometrically it means that you calculate a projection of the vector to
RyoheiHagimoto	0:0e0631af0305	2262	a subspace formed by a few eigenvectors corresponding to the dominant
RyoheiHagimoto	0:0e0631af0305	2263	eigenvalues of the covariance matrix. And usually such a projection is
RyoheiHagimoto	0:0e0631af0305	2264	very close to the original vector. So, you can represent the original
RyoheiHagimoto	0:0e0631af0305	2265	vector from a high-dimensional space with a much shorter vector
RyoheiHagimoto	0:0e0631af0305	2266	consisting of the projected vector's coordinates in the subspace. Such a
RyoheiHagimoto	0:0e0631af0305	2267	transformation is also known as Karhunen-Loeve Transform, or KLT.
RyoheiHagimoto	0:0e0631af0305	2268	See http://en.wikipedia.org/wiki/Principal_component_analysis
RyoheiHagimoto	0:0e0631af0305	2269
RyoheiHagimoto	0:0e0631af0305	2270	The sample below is the function that takes two matrices. The first
RyoheiHagimoto	0:0e0631af0305	2271	function stores a set of vectors (a row per vector) that is used to
RyoheiHagimoto	0:0e0631af0305	2272	calculate PCA. The second function stores another "test" set of vectors
RyoheiHagimoto	0:0e0631af0305	2273	(a row per vector). First, these vectors are compressed with PCA, then
RyoheiHagimoto	0:0e0631af0305	2274	reconstructed back, and then the reconstruction error norm is computed
RyoheiHagimoto	0:0e0631af0305	2275	and printed for each vector. :
RyoheiHagimoto	0:0e0631af0305	2276
RyoheiHagimoto	0:0e0631af0305	2277	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	2278	using namespace cv;
RyoheiHagimoto	0:0e0631af0305	2279
RyoheiHagimoto	0:0e0631af0305	2280	PCA compressPCA(const Mat& pcaset, int maxComponents,
RyoheiHagimoto	0:0e0631af0305	2281	const Mat& testset, Mat& compressed)
RyoheiHagimoto	0:0e0631af0305	2282	{
RyoheiHagimoto	0:0e0631af0305	2283	PCA pca(pcaset, // pass the data
RyoheiHagimoto	0:0e0631af0305	2284	Mat(), // we do not have a pre-computed mean vector,
RyoheiHagimoto	0:0e0631af0305	2285	// so let the PCA engine to compute it
RyoheiHagimoto	0:0e0631af0305	2286	PCA::DATA_AS_ROW, // indicate that the vectors
RyoheiHagimoto	0:0e0631af0305	2287	// are stored as matrix rows
RyoheiHagimoto	0:0e0631af0305	2288	// (use PCA::DATA_AS_COL if the vectors are
RyoheiHagimoto	0:0e0631af0305	2289	// the matrix columns)
RyoheiHagimoto	0:0e0631af0305	2290	maxComponents // specify, how many principal components to retain
RyoheiHagimoto	0:0e0631af0305	2291	);
RyoheiHagimoto	0:0e0631af0305	2292	// if there is no test data, just return the computed basis, ready-to-use
RyoheiHagimoto	0:0e0631af0305	2293	if( !testset.data )
RyoheiHagimoto	0:0e0631af0305	2294	return pca;
RyoheiHagimoto	0:0e0631af0305	2295	CV_Assert( testset.cols == pcaset.cols );
RyoheiHagimoto	0:0e0631af0305	2296
RyoheiHagimoto	0:0e0631af0305	2297	compressed.create(testset.rows, maxComponents, testset.type());
RyoheiHagimoto	0:0e0631af0305	2298
RyoheiHagimoto	0:0e0631af0305	2299	Mat reconstructed;
RyoheiHagimoto	0:0e0631af0305	2300	for( int i = 0; i < testset.rows; i++ )
RyoheiHagimoto	0:0e0631af0305	2301	{
RyoheiHagimoto	0:0e0631af0305	2302	Mat vec = testset.row(i), coeffs = compressed.row(i), reconstructed;
RyoheiHagimoto	0:0e0631af0305	2303	// compress the vector, the result will be stored
RyoheiHagimoto	0:0e0631af0305	2304	// in the i-th row of the output matrix
RyoheiHagimoto	0:0e0631af0305	2305	pca.project(vec, coeffs);
RyoheiHagimoto	0:0e0631af0305	2306	// and then reconstruct it
RyoheiHagimoto	0:0e0631af0305	2307	pca.backProject(coeffs, reconstructed);
RyoheiHagimoto	0:0e0631af0305	2308	// and measure the error
RyoheiHagimoto	0:0e0631af0305	2309	printf("%d. diff = %g\n", i, norm(vec, reconstructed, NORM_L2));
RyoheiHagimoto	0:0e0631af0305	2310	}
RyoheiHagimoto	0:0e0631af0305	2311	return pca;
RyoheiHagimoto	0:0e0631af0305	2312	}
RyoheiHagimoto	0:0e0631af0305	2313	@endcode
RyoheiHagimoto	0:0e0631af0305	2314	@sa calcCovarMatrix, mulTransposed, SVD, dft, dct
RyoheiHagimoto	0:0e0631af0305	2315	*/
RyoheiHagimoto	0:0e0631af0305	2316	class CV_EXPORTS PCA
RyoheiHagimoto	0:0e0631af0305	2317	{
RyoheiHagimoto	0:0e0631af0305	2318	public:
RyoheiHagimoto	0:0e0631af0305	2319	enum Flags { DATA_AS_ROW = 0, //!< indicates that the input samples are stored as matrix rows
RyoheiHagimoto	0:0e0631af0305	2320	DATA_AS_COL = 1, //!< indicates that the input samples are stored as matrix columns
RyoheiHagimoto	0:0e0631af0305	2321	USE_AVG = 2 //!
RyoheiHagimoto	0:0e0631af0305	2322	};
RyoheiHagimoto	0:0e0631af0305	2323
RyoheiHagimoto	0:0e0631af0305	2324	/** @brief default constructor
RyoheiHagimoto	0:0e0631af0305	2325
RyoheiHagimoto	0:0e0631af0305	2326	The default constructor initializes an empty %PCA structure. The other
RyoheiHagimoto	0:0e0631af0305	2327	constructors initialize the structure and call PCA::operator()().
RyoheiHagimoto	0:0e0631af0305	2328	*/
RyoheiHagimoto	0:0e0631af0305	2329	PCA();
RyoheiHagimoto	0:0e0631af0305	2330
RyoheiHagimoto	0:0e0631af0305	2331	/** @overload
RyoheiHagimoto	0:0e0631af0305	2332	@param data input samples stored as matrix rows or matrix columns.
RyoheiHagimoto	0:0e0631af0305	2333	@param mean optional mean value; if the matrix is empty (@c noArray()),
RyoheiHagimoto	0:0e0631af0305	2334	the mean is computed from the data.
RyoheiHagimoto	0:0e0631af0305	2335	@param flags operation flags; currently the parameter is only used to
RyoheiHagimoto	0:0e0631af0305	2336	specify the data layout (PCA::Flags)
RyoheiHagimoto	0:0e0631af0305	2337	@param maxComponents maximum number of components that %PCA should
RyoheiHagimoto	0:0e0631af0305	2338	retain; by default, all the components are retained.
RyoheiHagimoto	0:0e0631af0305	2339	*/
RyoheiHagimoto	0:0e0631af0305	2340	PCA(InputArray data, InputArray mean, int flags, int maxComponents = 0);
RyoheiHagimoto	0:0e0631af0305	2341
RyoheiHagimoto	0:0e0631af0305	2342	/** @overload
RyoheiHagimoto	0:0e0631af0305	2343	@param data input samples stored as matrix rows or matrix columns.
RyoheiHagimoto	0:0e0631af0305	2344	@param mean optional mean value; if the matrix is empty (noArray()),
RyoheiHagimoto	0:0e0631af0305	2345	the mean is computed from the data.
RyoheiHagimoto	0:0e0631af0305	2346	@param flags operation flags; currently the parameter is only used to
RyoheiHagimoto	0:0e0631af0305	2347	specify the data layout (PCA::Flags)
RyoheiHagimoto	0:0e0631af0305	2348	@param retainedVariance Percentage of variance that PCA should retain.
RyoheiHagimoto	0:0e0631af0305	2349	Using this parameter will let the PCA decided how many components to
RyoheiHagimoto	0:0e0631af0305	2350	retain but it will always keep at least 2.
RyoheiHagimoto	0:0e0631af0305	2351	*/
RyoheiHagimoto	0:0e0631af0305	2352	PCA(InputArray data, InputArray mean, int flags, double retainedVariance);
RyoheiHagimoto	0:0e0631af0305	2353
RyoheiHagimoto	0:0e0631af0305	2354	/** @brief performs %PCA
RyoheiHagimoto	0:0e0631af0305	2355
RyoheiHagimoto	0:0e0631af0305	2356	The operator performs %PCA of the supplied dataset. It is safe to reuse
RyoheiHagimoto	0:0e0631af0305	2357	the same PCA structure for multiple datasets. That is, if the structure
RyoheiHagimoto	0:0e0631af0305	2358	has been previously used with another dataset, the existing internal
RyoheiHagimoto	0:0e0631af0305	2359	data is reclaimed and the new @ref eigenvalues, @ref eigenvectors and @ref
RyoheiHagimoto	0:0e0631af0305	2360	mean are allocated and computed.
RyoheiHagimoto	0:0e0631af0305	2361
RyoheiHagimoto	0:0e0631af0305	2362	The computed @ref eigenvalues are sorted from the largest to the smallest and
RyoheiHagimoto	0:0e0631af0305	2363	the corresponding @ref eigenvectors are stored as eigenvectors rows.
RyoheiHagimoto	0:0e0631af0305	2364
RyoheiHagimoto	0:0e0631af0305	2365	@param data input samples stored as the matrix rows or as the matrix
RyoheiHagimoto	0:0e0631af0305	2366	columns.
RyoheiHagimoto	0:0e0631af0305	2367	@param mean optional mean value; if the matrix is empty (noArray()),
RyoheiHagimoto	0:0e0631af0305	2368	the mean is computed from the data.
RyoheiHagimoto	0:0e0631af0305	2369	@param flags operation flags; currently the parameter is only used to
RyoheiHagimoto	0:0e0631af0305	2370	specify the data layout. (Flags)
RyoheiHagimoto	0:0e0631af0305	2371	@param maxComponents maximum number of components that PCA should
RyoheiHagimoto	0:0e0631af0305	2372	retain; by default, all the components are retained.
RyoheiHagimoto	0:0e0631af0305	2373	*/
RyoheiHagimoto	0:0e0631af0305	2374	PCA& operator()(InputArray data, InputArray mean, int flags, int maxComponents = 0);
RyoheiHagimoto	0:0e0631af0305	2375
RyoheiHagimoto	0:0e0631af0305	2376	/** @overload
RyoheiHagimoto	0:0e0631af0305	2377	@param data input samples stored as the matrix rows or as the matrix
RyoheiHagimoto	0:0e0631af0305	2378	columns.
RyoheiHagimoto	0:0e0631af0305	2379	@param mean optional mean value; if the matrix is empty (noArray()),
RyoheiHagimoto	0:0e0631af0305	2380	the mean is computed from the data.
RyoheiHagimoto	0:0e0631af0305	2381	@param flags operation flags; currently the parameter is only used to
RyoheiHagimoto	0:0e0631af0305	2382	specify the data layout. (PCA::Flags)
RyoheiHagimoto	0:0e0631af0305	2383	@param retainedVariance Percentage of variance that %PCA should retain.
RyoheiHagimoto	0:0e0631af0305	2384	Using this parameter will let the %PCA decided how many components to
RyoheiHagimoto	0:0e0631af0305	2385	retain but it will always keep at least 2.
RyoheiHagimoto	0:0e0631af0305	2386	*/
RyoheiHagimoto	0:0e0631af0305	2387	PCA& operator()(InputArray data, InputArray mean, int flags, double retainedVariance);
RyoheiHagimoto	0:0e0631af0305	2388
RyoheiHagimoto	0:0e0631af0305	2389	/** @brief Projects vector(s) to the principal component subspace.
RyoheiHagimoto	0:0e0631af0305	2390
RyoheiHagimoto	0:0e0631af0305	2391	The methods project one or more vectors to the principal component
RyoheiHagimoto	0:0e0631af0305	2392	subspace, where each vector projection is represented by coefficients in
RyoheiHagimoto	0:0e0631af0305	2393	the principal component basis. The first form of the method returns the
RyoheiHagimoto	0:0e0631af0305	2394	matrix that the second form writes to the result. So the first form can
RyoheiHagimoto	0:0e0631af0305	2395	be used as a part of expression while the second form can be more
RyoheiHagimoto	0:0e0631af0305	2396	efficient in a processing loop.
RyoheiHagimoto	0:0e0631af0305	2397	@param vec input vector(s); must have the same dimensionality and the
RyoheiHagimoto	0:0e0631af0305	2398	same layout as the input data used at %PCA phase, that is, if
RyoheiHagimoto	0:0e0631af0305	2399	DATA_AS_ROW are specified, then `vec.cols==data.cols`
RyoheiHagimoto	0:0e0631af0305	2400	(vector dimensionality) and `vec.rows` is the number of vectors to
RyoheiHagimoto	0:0e0631af0305	2401	project, and the same is true for the PCA::DATA_AS_COL case.
RyoheiHagimoto	0:0e0631af0305	2402	*/
RyoheiHagimoto	0:0e0631af0305	2403	Mat project(InputArray vec) const;
RyoheiHagimoto	0:0e0631af0305	2404
RyoheiHagimoto	0:0e0631af0305	2405	/** @overload
RyoheiHagimoto	0:0e0631af0305	2406	@param vec input vector(s); must have the same dimensionality and the
RyoheiHagimoto	0:0e0631af0305	2407	same layout as the input data used at PCA phase, that is, if
RyoheiHagimoto	0:0e0631af0305	2408	DATA_AS_ROW are specified, then `vec.cols==data.cols`
RyoheiHagimoto	0:0e0631af0305	2409	(vector dimensionality) and `vec.rows` is the number of vectors to
RyoheiHagimoto	0:0e0631af0305	2410	project, and the same is true for the PCA::DATA_AS_COL case.
RyoheiHagimoto	0:0e0631af0305	2411	@param result output vectors; in case of PCA::DATA_AS_COL, the
RyoheiHagimoto	0:0e0631af0305	2412	output matrix has as many columns as the number of input vectors, this
RyoheiHagimoto	0:0e0631af0305	2413	means that `result.cols==vec.cols` and the number of rows match the
RyoheiHagimoto	0:0e0631af0305	2414	number of principal components (for example, `maxComponents` parameter
RyoheiHagimoto	0:0e0631af0305	2415	passed to the constructor).
RyoheiHagimoto	0:0e0631af0305	2416	*/
RyoheiHagimoto	0:0e0631af0305	2417	void project(InputArray vec, OutputArray result) const;
RyoheiHagimoto	0:0e0631af0305	2418
RyoheiHagimoto	0:0e0631af0305	2419	/** @brief Reconstructs vectors from their PC projections.
RyoheiHagimoto	0:0e0631af0305	2420
RyoheiHagimoto	0:0e0631af0305	2421	The methods are inverse operations to PCA::project. They take PC
RyoheiHagimoto	0:0e0631af0305	2422	coordinates of projected vectors and reconstruct the original vectors.
RyoheiHagimoto	0:0e0631af0305	2423	Unless all the principal components have been retained, the
RyoheiHagimoto	0:0e0631af0305	2424	reconstructed vectors are different from the originals. But typically,
RyoheiHagimoto	0:0e0631af0305	2425	the difference is small if the number of components is large enough (but
RyoheiHagimoto	0:0e0631af0305	2426	still much smaller than the original vector dimensionality). As a
RyoheiHagimoto	0:0e0631af0305	2427	result, PCA is used.
RyoheiHagimoto	0:0e0631af0305	2428	@param vec coordinates of the vectors in the principal component
RyoheiHagimoto	0:0e0631af0305	2429	subspace, the layout and size are the same as of PCA::project output
RyoheiHagimoto	0:0e0631af0305	2430	vectors.
RyoheiHagimoto	0:0e0631af0305	2431	*/
RyoheiHagimoto	0:0e0631af0305	2432	Mat backProject(InputArray vec) const;
RyoheiHagimoto	0:0e0631af0305	2433
RyoheiHagimoto	0:0e0631af0305	2434	/** @overload
RyoheiHagimoto	0:0e0631af0305	2435	@param vec coordinates of the vectors in the principal component
RyoheiHagimoto	0:0e0631af0305	2436	subspace, the layout and size are the same as of PCA::project output
RyoheiHagimoto	0:0e0631af0305	2437	vectors.
RyoheiHagimoto	0:0e0631af0305	2438	@param result reconstructed vectors; the layout and size are the same as
RyoheiHagimoto	0:0e0631af0305	2439	of PCA::project input vectors.
RyoheiHagimoto	0:0e0631af0305	2440	*/
RyoheiHagimoto	0:0e0631af0305	2441	void backProject(InputArray vec, OutputArray result) const;
RyoheiHagimoto	0:0e0631af0305	2442
RyoheiHagimoto	0:0e0631af0305	2443	/** @brief write PCA objects
RyoheiHagimoto	0:0e0631af0305	2444
RyoheiHagimoto	0:0e0631af0305	2445	Writes @ref eigenvalues @ref eigenvectors and @ref mean to specified FileStorage
RyoheiHagimoto	0:0e0631af0305	2446	*/
RyoheiHagimoto	0:0e0631af0305	2447	void write(FileStorage& fs) const;
RyoheiHagimoto	0:0e0631af0305	2448
RyoheiHagimoto	0:0e0631af0305	2449	/** @brief load PCA objects
RyoheiHagimoto	0:0e0631af0305	2450
RyoheiHagimoto	0:0e0631af0305	2451	Loads @ref eigenvalues @ref eigenvectors and @ref mean from specified FileNode
RyoheiHagimoto	0:0e0631af0305	2452	*/
RyoheiHagimoto	0:0e0631af0305	2453	void read(const FileNode& fn);
RyoheiHagimoto	0:0e0631af0305	2454
RyoheiHagimoto	0:0e0631af0305	2455	Mat eigenvectors; //!< eigenvectors of the covariation matrix
RyoheiHagimoto	0:0e0631af0305	2456	Mat eigenvalues; //!< eigenvalues of the covariation matrix
RyoheiHagimoto	0:0e0631af0305	2457	Mat mean; //!< mean value subtracted before the projection and added after the back projection
RyoheiHagimoto	0:0e0631af0305	2458	};
RyoheiHagimoto	0:0e0631af0305	2459
RyoheiHagimoto	0:0e0631af0305	2460	/** @example pca.cpp
RyoheiHagimoto	0:0e0631af0305	2461	An example using %PCA for dimensionality reduction while maintaining an amount of variance
RyoheiHagimoto	0:0e0631af0305	2462	*/
RyoheiHagimoto	0:0e0631af0305	2463
RyoheiHagimoto	0:0e0631af0305	2464	/**
RyoheiHagimoto	0:0e0631af0305	2465	@brief Linear Discriminant Analysis
RyoheiHagimoto	0:0e0631af0305	2466	@todo document this class
RyoheiHagimoto	0:0e0631af0305	2467	*/
RyoheiHagimoto	0:0e0631af0305	2468	class CV_EXPORTS LDA
RyoheiHagimoto	0:0e0631af0305	2469	{
RyoheiHagimoto	0:0e0631af0305	2470	public:
RyoheiHagimoto	0:0e0631af0305	2471	/** @brief constructor
RyoheiHagimoto	0:0e0631af0305	2472	Initializes a LDA with num_components (default 0).
RyoheiHagimoto	0:0e0631af0305	2473	*/
RyoheiHagimoto	0:0e0631af0305	2474	explicit LDA(int num_components = 0);
RyoheiHagimoto	0:0e0631af0305	2475
RyoheiHagimoto	0:0e0631af0305	2476	/** Initializes and performs a Discriminant Analysis with Fisher's
RyoheiHagimoto	0:0e0631af0305	2477	Optimization Criterion on given data in src and corresponding labels
RyoheiHagimoto	0:0e0631af0305	2478	in labels. If 0 (or less) number of components are given, they are
RyoheiHagimoto	0:0e0631af0305	2479	automatically determined for given data in computation.
RyoheiHagimoto	0:0e0631af0305	2480	*/
RyoheiHagimoto	0:0e0631af0305	2481	LDA(InputArrayOfArrays src, InputArray labels, int num_components = 0);
RyoheiHagimoto	0:0e0631af0305	2482
RyoheiHagimoto	0:0e0631af0305	2483	/** Serializes this object to a given filename.
RyoheiHagimoto	0:0e0631af0305	2484	*/
RyoheiHagimoto	0:0e0631af0305	2485	void save(const String& filename) const;
RyoheiHagimoto	0:0e0631af0305	2486
RyoheiHagimoto	0:0e0631af0305	2487	/** Deserializes this object from a given filename.
RyoheiHagimoto	0:0e0631af0305	2488	*/
RyoheiHagimoto	0:0e0631af0305	2489	void load(const String& filename);
RyoheiHagimoto	0:0e0631af0305	2490
RyoheiHagimoto	0:0e0631af0305	2491	/** Serializes this object to a given cv::FileStorage.
RyoheiHagimoto	0:0e0631af0305	2492	*/
RyoheiHagimoto	0:0e0631af0305	2493	void save(FileStorage& fs) const;
RyoheiHagimoto	0:0e0631af0305	2494
RyoheiHagimoto	0:0e0631af0305	2495	/** Deserializes this object from a given cv::FileStorage.
RyoheiHagimoto	0:0e0631af0305	2496	*/
RyoheiHagimoto	0:0e0631af0305	2497	void load(const FileStorage& node);
RyoheiHagimoto	0:0e0631af0305	2498
RyoheiHagimoto	0:0e0631af0305	2499	/** destructor
RyoheiHagimoto	0:0e0631af0305	2500	*/
RyoheiHagimoto	0:0e0631af0305	2501	~LDA();
RyoheiHagimoto	0:0e0631af0305	2502
RyoheiHagimoto	0:0e0631af0305	2503	/** Compute the discriminants for data in src (row aligned) and labels.
RyoheiHagimoto	0:0e0631af0305	2504	*/
RyoheiHagimoto	0:0e0631af0305	2505	void compute(InputArrayOfArrays src, InputArray labels);
RyoheiHagimoto	0:0e0631af0305	2506
RyoheiHagimoto	0:0e0631af0305	2507	/** Projects samples into the LDA subspace.
RyoheiHagimoto	0:0e0631af0305	2508	src may be one or more row aligned samples.
RyoheiHagimoto	0:0e0631af0305	2509	*/
RyoheiHagimoto	0:0e0631af0305	2510	Mat project(InputArray src);
RyoheiHagimoto	0:0e0631af0305	2511
RyoheiHagimoto	0:0e0631af0305	2512	/** Reconstructs projections from the LDA subspace.
RyoheiHagimoto	0:0e0631af0305	2513	src may be one or more row aligned projections.
RyoheiHagimoto	0:0e0631af0305	2514	*/
RyoheiHagimoto	0:0e0631af0305	2515	Mat reconstruct(InputArray src);
RyoheiHagimoto	0:0e0631af0305	2516
RyoheiHagimoto	0:0e0631af0305	2517	/** Returns the eigenvectors of this LDA.
RyoheiHagimoto	0:0e0631af0305	2518	*/
RyoheiHagimoto	0:0e0631af0305	2519	Mat eigenvectors() const { return _eigenvectors; }
RyoheiHagimoto	0:0e0631af0305	2520
RyoheiHagimoto	0:0e0631af0305	2521	/** Returns the eigenvalues of this LDA.
RyoheiHagimoto	0:0e0631af0305	2522	*/
RyoheiHagimoto	0:0e0631af0305	2523	Mat eigenvalues() const { return _eigenvalues; }
RyoheiHagimoto	0:0e0631af0305	2524
RyoheiHagimoto	0:0e0631af0305	2525	static Mat subspaceProject(InputArray W, InputArray mean, InputArray src);
RyoheiHagimoto	0:0e0631af0305	2526	static Mat subspaceReconstruct(InputArray W, InputArray mean, InputArray src);
RyoheiHagimoto	0:0e0631af0305	2527
RyoheiHagimoto	0:0e0631af0305	2528	protected:
RyoheiHagimoto	0:0e0631af0305	2529	bool _dataAsRow; // unused, but needed for 3.0 ABI compatibility.
RyoheiHagimoto	0:0e0631af0305	2530	int _num_components;
RyoheiHagimoto	0:0e0631af0305	2531	Mat _eigenvectors;
RyoheiHagimoto	0:0e0631af0305	2532	Mat _eigenvalues;
RyoheiHagimoto	0:0e0631af0305	2533	void lda(InputArrayOfArrays src, InputArray labels);
RyoheiHagimoto	0:0e0631af0305	2534	};
RyoheiHagimoto	0:0e0631af0305	2535
RyoheiHagimoto	0:0e0631af0305	2536	/** @brief Singular Value Decomposition
RyoheiHagimoto	0:0e0631af0305	2537
RyoheiHagimoto	0:0e0631af0305	2538	Class for computing Singular Value Decomposition of a floating-point
RyoheiHagimoto	0:0e0631af0305	2539	matrix. The Singular Value Decomposition is used to solve least-square
RyoheiHagimoto	0:0e0631af0305	2540	problems, under-determined linear systems, invert matrices, compute
RyoheiHagimoto	0:0e0631af0305	2541	condition numbers, and so on.
RyoheiHagimoto	0:0e0631af0305	2542
RyoheiHagimoto	0:0e0631af0305	2543	If you want to compute a condition number of a matrix or an absolute value of
RyoheiHagimoto	0:0e0631af0305	2544	its determinant, you do not need `u` and `vt`. You can pass
RyoheiHagimoto	0:0e0631af0305	2545	flags=SVD::NO_UV\|... . Another flag SVD::FULL_UV indicates that full-size u
RyoheiHagimoto	0:0e0631af0305	2546	and vt must be computed, which is not necessary most of the time.
RyoheiHagimoto	0:0e0631af0305	2547
RyoheiHagimoto	0:0e0631af0305	2548	@sa invert, solve, eigen, determinant
RyoheiHagimoto	0:0e0631af0305	2549	*/
RyoheiHagimoto	0:0e0631af0305	2550	class CV_EXPORTS SVD
RyoheiHagimoto	0:0e0631af0305	2551	{
RyoheiHagimoto	0:0e0631af0305	2552	public:
RyoheiHagimoto	0:0e0631af0305	2553	enum Flags {
RyoheiHagimoto	0:0e0631af0305	2554	/** allow the algorithm to modify the decomposed matrix; it can save space and speed up
RyoheiHagimoto	0:0e0631af0305	2555	processing. currently ignored. */
RyoheiHagimoto	0:0e0631af0305	2556	MODIFY_A = 1,
RyoheiHagimoto	0:0e0631af0305	2557	/** indicates that only a vector of singular values `w` is to be processed, while u and vt
RyoheiHagimoto	0:0e0631af0305	2558	will be set to empty matrices */
RyoheiHagimoto	0:0e0631af0305	2559	NO_UV = 2,
RyoheiHagimoto	0:0e0631af0305	2560	/** when the matrix is not square, by default the algorithm produces u and vt matrices of
RyoheiHagimoto	0:0e0631af0305	2561	sufficiently large size for the further A reconstruction; if, however, FULL_UV flag is
RyoheiHagimoto	0:0e0631af0305	2562	specified, u and vt will be full-size square orthogonal matrices.*/
RyoheiHagimoto	0:0e0631af0305	2563	FULL_UV = 4
RyoheiHagimoto	0:0e0631af0305	2564	};
RyoheiHagimoto	0:0e0631af0305	2565
RyoheiHagimoto	0:0e0631af0305	2566	/** @brief the default constructor
RyoheiHagimoto	0:0e0631af0305	2567
RyoheiHagimoto	0:0e0631af0305	2568	initializes an empty SVD structure
RyoheiHagimoto	0:0e0631af0305	2569	*/
RyoheiHagimoto	0:0e0631af0305	2570	SVD();
RyoheiHagimoto	0:0e0631af0305	2571
RyoheiHagimoto	0:0e0631af0305	2572	/** @overload
RyoheiHagimoto	0:0e0631af0305	2573	initializes an empty SVD structure and then calls SVD::operator()
RyoheiHagimoto	0:0e0631af0305	2574	@param src decomposed matrix.
RyoheiHagimoto	0:0e0631af0305	2575	@param flags operation flags (SVD::Flags)
RyoheiHagimoto	0:0e0631af0305	2576	*/
RyoheiHagimoto	0:0e0631af0305	2577	SVD( InputArray src, int flags = 0 );
RyoheiHagimoto	0:0e0631af0305	2578
RyoheiHagimoto	0:0e0631af0305	2579	/** @brief the operator that performs SVD. The previously allocated u, w and vt are released.
RyoheiHagimoto	0:0e0631af0305	2580
RyoheiHagimoto	0:0e0631af0305	2581	The operator performs the singular value decomposition of the supplied
RyoheiHagimoto	0:0e0631af0305	2582	matrix. The u,`vt` , and the vector of singular values w are stored in
RyoheiHagimoto	0:0e0631af0305	2583	the structure. The same SVD structure can be reused many times with
RyoheiHagimoto	0:0e0631af0305	2584	different matrices. Each time, if needed, the previous u,`vt` , and w
RyoheiHagimoto	0:0e0631af0305	2585	are reclaimed and the new matrices are created, which is all handled by
RyoheiHagimoto	0:0e0631af0305	2586	Mat::create.
RyoheiHagimoto	0:0e0631af0305	2587	@param src decomposed matrix.
RyoheiHagimoto	0:0e0631af0305	2588	@param flags operation flags (SVD::Flags)
RyoheiHagimoto	0:0e0631af0305	2589	*/
RyoheiHagimoto	0:0e0631af0305	2590	SVD& operator ()( InputArray src, int flags = 0 );
RyoheiHagimoto	0:0e0631af0305	2591
RyoheiHagimoto	0:0e0631af0305	2592	/** @brief decomposes matrix and stores the results to user-provided matrices
RyoheiHagimoto	0:0e0631af0305	2593
RyoheiHagimoto	0:0e0631af0305	2594	The methods/functions perform SVD of matrix. Unlike SVD::SVD constructor
RyoheiHagimoto	0:0e0631af0305	2595	and SVD::operator(), they store the results to the user-provided
RyoheiHagimoto	0:0e0631af0305	2596	matrices:
RyoheiHagimoto	0:0e0631af0305	2597
RyoheiHagimoto	0:0e0631af0305	2598	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	2599	Mat A, w, u, vt;
RyoheiHagimoto	0:0e0631af0305	2600	SVD::compute(A, w, u, vt);
RyoheiHagimoto	0:0e0631af0305	2601	@endcode
RyoheiHagimoto	0:0e0631af0305	2602
RyoheiHagimoto	0:0e0631af0305	2603	@param src decomposed matrix
RyoheiHagimoto	0:0e0631af0305	2604	@param w calculated singular values
RyoheiHagimoto	0:0e0631af0305	2605	@param u calculated left singular vectors
RyoheiHagimoto	0:0e0631af0305	2606	@param vt transposed matrix of right singular values
RyoheiHagimoto	0:0e0631af0305	2607	@param flags operation flags - see SVD::SVD.
RyoheiHagimoto	0:0e0631af0305	2608	*/
RyoheiHagimoto	0:0e0631af0305	2609	static void compute( InputArray src, OutputArray w,
RyoheiHagimoto	0:0e0631af0305	2610	OutputArray u, OutputArray vt, int flags = 0 );
RyoheiHagimoto	0:0e0631af0305	2611
RyoheiHagimoto	0:0e0631af0305	2612	/** @overload
RyoheiHagimoto	0:0e0631af0305	2613	computes singular values of a matrix
RyoheiHagimoto	0:0e0631af0305	2614	@param src decomposed matrix
RyoheiHagimoto	0:0e0631af0305	2615	@param w calculated singular values
RyoheiHagimoto	0:0e0631af0305	2616	@param flags operation flags - see SVD::Flags.
RyoheiHagimoto	0:0e0631af0305	2617	*/
RyoheiHagimoto	0:0e0631af0305	2618	static void compute( InputArray src, OutputArray w, int flags = 0 );
RyoheiHagimoto	0:0e0631af0305	2619
RyoheiHagimoto	0:0e0631af0305	2620	/** @brief performs back substitution
RyoheiHagimoto	0:0e0631af0305	2621	*/
RyoheiHagimoto	0:0e0631af0305	2622	static void backSubst( InputArray w, InputArray u,
RyoheiHagimoto	0:0e0631af0305	2623	InputArray vt, InputArray rhs,
RyoheiHagimoto	0:0e0631af0305	2624	OutputArray dst );
RyoheiHagimoto	0:0e0631af0305	2625
RyoheiHagimoto	0:0e0631af0305	2626	/** @brief solves an under-determined singular linear system
RyoheiHagimoto	0:0e0631af0305	2627
RyoheiHagimoto	0:0e0631af0305	2628	The method finds a unit-length solution x of a singular linear system
RyoheiHagimoto	0:0e0631af0305	2629	A\*x = 0. Depending on the rank of A, there can be no solutions, a
RyoheiHagimoto	0:0e0631af0305	2630	single solution or an infinite number of solutions. In general, the
RyoheiHagimoto	0:0e0631af0305	2631	algorithm solves the following problem:
RyoheiHagimoto	0:0e0631af0305	2632	\f[dst = \arg \min _{x: \\| x \\| =1} \\| src \cdot x \\|\f]
RyoheiHagimoto	0:0e0631af0305	2633	@param src left-hand-side matrix.
RyoheiHagimoto	0:0e0631af0305	2634	@param dst found solution.
RyoheiHagimoto	0:0e0631af0305	2635	*/
RyoheiHagimoto	0:0e0631af0305	2636	static void solveZ( InputArray src, OutputArray dst );
RyoheiHagimoto	0:0e0631af0305	2637
RyoheiHagimoto	0:0e0631af0305	2638	/** @brief performs a singular value back substitution.
RyoheiHagimoto	0:0e0631af0305	2639
RyoheiHagimoto	0:0e0631af0305	2640	The method calculates a back substitution for the specified right-hand
RyoheiHagimoto	0:0e0631af0305	2641	side:
RyoheiHagimoto	0:0e0631af0305	2642
RyoheiHagimoto	0:0e0631af0305	2643	\f[\texttt{x} = \texttt{vt} ^T \cdot diag( \texttt{w} )^{-1} \cdot \texttt{u} ^T \cdot \texttt{rhs} \sim \texttt{A} ^{-1} \cdot \texttt{rhs}\f]
RyoheiHagimoto	0:0e0631af0305	2644
RyoheiHagimoto	0:0e0631af0305	2645	Using this technique you can either get a very accurate solution of the
RyoheiHagimoto	0:0e0631af0305	2646	convenient linear system, or the best (in the least-squares terms)
RyoheiHagimoto	0:0e0631af0305	2647	pseudo-solution of an overdetermined linear system.
RyoheiHagimoto	0:0e0631af0305	2648
RyoheiHagimoto	0:0e0631af0305	2649	@param rhs right-hand side of a linear system (u\w\v')\*dst = rhs to
RyoheiHagimoto	0:0e0631af0305	2650	be solved, where A has been previously decomposed.
RyoheiHagimoto	0:0e0631af0305	2651
RyoheiHagimoto	0:0e0631af0305	2652	@param dst found solution of the system.
RyoheiHagimoto	0:0e0631af0305	2653
RyoheiHagimoto	0:0e0631af0305	2654	@note Explicit SVD with the further back substitution only makes sense
RyoheiHagimoto	0:0e0631af0305	2655	if you need to solve many linear systems with the same left-hand side
RyoheiHagimoto	0:0e0631af0305	2656	(for example, src ). If all you need is to solve a single system
RyoheiHagimoto	0:0e0631af0305	2657	(possibly with multiple rhs immediately available), simply call solve
RyoheiHagimoto	0:0e0631af0305	2658	add pass DECOMP_SVD there. It does absolutely the same thing.
RyoheiHagimoto	0:0e0631af0305	2659	*/
RyoheiHagimoto	0:0e0631af0305	2660	void backSubst( InputArray rhs, OutputArray dst ) const;
RyoheiHagimoto	0:0e0631af0305	2661
RyoheiHagimoto	0:0e0631af0305	2662	/** @todo document */
RyoheiHagimoto	0:0e0631af0305	2663	template<typename _Tp, int m, int n, int nm> static
RyoheiHagimoto	0:0e0631af0305	2664	void compute( const Matx<_Tp, m, n>& a, Matx<_Tp, nm, 1>& w, Matx<_Tp, m, nm>& u, Matx<_Tp, n, nm>& vt );
RyoheiHagimoto	0:0e0631af0305	2665
RyoheiHagimoto	0:0e0631af0305	2666	/** @todo document */
RyoheiHagimoto	0:0e0631af0305	2667	template<typename _Tp, int m, int n, int nm> static
RyoheiHagimoto	0:0e0631af0305	2668	void compute( const Matx<_Tp, m, n>& a, Matx<_Tp, nm, 1>& w );
RyoheiHagimoto	0:0e0631af0305	2669
RyoheiHagimoto	0:0e0631af0305	2670	/** @todo document */
RyoheiHagimoto	0:0e0631af0305	2671	template<typename _Tp, int m, int n, int nm, int nb> static
RyoheiHagimoto	0:0e0631af0305	2672	void backSubst( const Matx<_Tp, nm, 1>& w, const Matx<_Tp, m, nm>& u, const Matx<_Tp, n, nm>& vt, const Matx<_Tp, m, nb>& rhs, Matx<_Tp, n, nb>& dst );
RyoheiHagimoto	0:0e0631af0305	2673
RyoheiHagimoto	0:0e0631af0305	2674	Mat u, w, vt;
RyoheiHagimoto	0:0e0631af0305	2675	};
RyoheiHagimoto	0:0e0631af0305	2676
RyoheiHagimoto	0:0e0631af0305	2677	/** @brief Random Number Generator
RyoheiHagimoto	0:0e0631af0305	2678
RyoheiHagimoto	0:0e0631af0305	2679	Random number generator. It encapsulates the state (currently, a 64-bit
RyoheiHagimoto	0:0e0631af0305	2680	integer) and has methods to return scalar random values and to fill
RyoheiHagimoto	0:0e0631af0305	2681	arrays with random values. Currently it supports uniform and Gaussian
RyoheiHagimoto	0:0e0631af0305	2682	(normal) distributions. The generator uses Multiply-With-Carry
RyoheiHagimoto	0:0e0631af0305	2683	algorithm, introduced by G. Marsaglia (
RyoheiHagimoto	0:0e0631af0305	2684	<http://en.wikipedia.org/wiki/Multiply-with-carry> ).
RyoheiHagimoto	0:0e0631af0305	2685	Gaussian-distribution random numbers are generated using the Ziggurat
RyoheiHagimoto	0:0e0631af0305	2686	algorithm ( <http://en.wikipedia.org/wiki/Ziggurat_algorithm> ),
RyoheiHagimoto	0:0e0631af0305	2687	introduced by G. Marsaglia and W. W. Tsang.
RyoheiHagimoto	0:0e0631af0305	2688	*/
RyoheiHagimoto	0:0e0631af0305	2689	class CV_EXPORTS RNG
RyoheiHagimoto	0:0e0631af0305	2690	{
RyoheiHagimoto	0:0e0631af0305	2691	public:
RyoheiHagimoto	0:0e0631af0305	2692	enum { UNIFORM = 0,
RyoheiHagimoto	0:0e0631af0305	2693	NORMAL = 1
RyoheiHagimoto	0:0e0631af0305	2694	};
RyoheiHagimoto	0:0e0631af0305	2695
RyoheiHagimoto	0:0e0631af0305	2696	/** @brief constructor
RyoheiHagimoto	0:0e0631af0305	2697
RyoheiHagimoto	0:0e0631af0305	2698	These are the RNG constructors. The first form sets the state to some
RyoheiHagimoto	0:0e0631af0305	2699	pre-defined value, equal to 2\\32-1 in the current implementation. The
RyoheiHagimoto	0:0e0631af0305	2700	second form sets the state to the specified value. If you passed state=0
RyoheiHagimoto	0:0e0631af0305	2701	, the constructor uses the above default value instead to avoid the
RyoheiHagimoto	0:0e0631af0305	2702	singular random number sequence, consisting of all zeros.
RyoheiHagimoto	0:0e0631af0305	2703	*/
RyoheiHagimoto	0:0e0631af0305	2704	RNG();
RyoheiHagimoto	0:0e0631af0305	2705	/** @overload
RyoheiHagimoto	0:0e0631af0305	2706	@param state 64-bit value used to initialize the RNG.
RyoheiHagimoto	0:0e0631af0305	2707	*/
RyoheiHagimoto	0:0e0631af0305	2708	RNG(uint64 state);
RyoheiHagimoto	0:0e0631af0305	2709	/**The method updates the state using the MWC algorithm and returns the
RyoheiHagimoto	0:0e0631af0305	2710	next 32-bit random number.*/
RyoheiHagimoto	0:0e0631af0305	2711	unsigned next();
RyoheiHagimoto	0:0e0631af0305	2712
RyoheiHagimoto	0:0e0631af0305	2713	/**Each of the methods updates the state using the MWC algorithm and
RyoheiHagimoto	0:0e0631af0305	2714	returns the next random number of the specified type. In case of integer
RyoheiHagimoto	0:0e0631af0305	2715	types, the returned number is from the available value range for the
RyoheiHagimoto	0:0e0631af0305	2716	specified type. In case of floating-point types, the returned value is
RyoheiHagimoto	0:0e0631af0305	2717	from [0,1) range.
RyoheiHagimoto	0:0e0631af0305	2718	*/
RyoheiHagimoto	0:0e0631af0305	2719	operator uchar();
RyoheiHagimoto	0:0e0631af0305	2720	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2721	operator schar();
RyoheiHagimoto	0:0e0631af0305	2722	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2723	operator ushort();
RyoheiHagimoto	0:0e0631af0305	2724	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2725	operator short();
RyoheiHagimoto	0:0e0631af0305	2726	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2727	operator unsigned();
RyoheiHagimoto	0:0e0631af0305	2728	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2729	operator int();
RyoheiHagimoto	0:0e0631af0305	2730	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2731	operator float();
RyoheiHagimoto	0:0e0631af0305	2732	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2733	operator double();
RyoheiHagimoto	0:0e0631af0305	2734
RyoheiHagimoto	0:0e0631af0305	2735	/** @brief returns a random integer sampled uniformly from [0, N).
RyoheiHagimoto	0:0e0631af0305	2736
RyoheiHagimoto	0:0e0631af0305	2737	The methods transform the state using the MWC algorithm and return the
RyoheiHagimoto	0:0e0631af0305	2738	next random number. The first form is equivalent to RNG::next . The
RyoheiHagimoto	0:0e0631af0305	2739	second form returns the random number modulo N , which means that the
RyoheiHagimoto	0:0e0631af0305	2740	result is in the range [0, N) .
RyoheiHagimoto	0:0e0631af0305	2741	*/
RyoheiHagimoto	0:0e0631af0305	2742	unsigned operator ()();
RyoheiHagimoto	0:0e0631af0305	2743	/** @overload
RyoheiHagimoto	0:0e0631af0305	2744	@param N upper non-inclusive boundary of the returned random number.
RyoheiHagimoto	0:0e0631af0305	2745	*/
RyoheiHagimoto	0:0e0631af0305	2746	unsigned operator ()(unsigned N);
RyoheiHagimoto	0:0e0631af0305	2747
RyoheiHagimoto	0:0e0631af0305	2748	/** @brief returns uniformly distributed integer random number from [a,b) range
RyoheiHagimoto	0:0e0631af0305	2749
RyoheiHagimoto	0:0e0631af0305	2750	The methods transform the state using the MWC algorithm and return the
RyoheiHagimoto	0:0e0631af0305	2751	next uniformly-distributed random number of the specified type, deduced
RyoheiHagimoto	0:0e0631af0305	2752	from the input parameter type, from the range [a, b) . There is a nuance
RyoheiHagimoto	0:0e0631af0305	2753	illustrated by the following sample:
RyoheiHagimoto	0:0e0631af0305	2754
RyoheiHagimoto	0:0e0631af0305	2755	@code{.cpp}
RyoheiHagimoto	0:0e0631af0305	2756	RNG rng;
RyoheiHagimoto	0:0e0631af0305	2757
RyoheiHagimoto	0:0e0631af0305	2758	// always produces 0
RyoheiHagimoto	0:0e0631af0305	2759	double a = rng.uniform(0, 1);
RyoheiHagimoto	0:0e0631af0305	2760
RyoheiHagimoto	0:0e0631af0305	2761	// produces double from [0, 1)
RyoheiHagimoto	0:0e0631af0305	2762	double a1 = rng.uniform((double)0, (double)1);
RyoheiHagimoto	0:0e0631af0305	2763
RyoheiHagimoto	0:0e0631af0305	2764	// produces float from [0, 1)
RyoheiHagimoto	0:0e0631af0305	2765	double b = rng.uniform(0.f, 1.f);
RyoheiHagimoto	0:0e0631af0305	2766
RyoheiHagimoto	0:0e0631af0305	2767	// produces double from [0, 1)
RyoheiHagimoto	0:0e0631af0305	2768	double c = rng.uniform(0., 1.);
RyoheiHagimoto	0:0e0631af0305	2769
RyoheiHagimoto	0:0e0631af0305	2770	// may cause compiler error because of ambiguity:
RyoheiHagimoto	0:0e0631af0305	2771	// RNG::uniform(0, (int)0.999999)? or RNG::uniform((double)0, 0.99999)?
RyoheiHagimoto	0:0e0631af0305	2772	double d = rng.uniform(0, 0.999999);
RyoheiHagimoto	0:0e0631af0305	2773	@endcode
RyoheiHagimoto	0:0e0631af0305	2774
RyoheiHagimoto	0:0e0631af0305	2775	The compiler does not take into account the type of the variable to
RyoheiHagimoto	0:0e0631af0305	2776	which you assign the result of RNG::uniform . The only thing that
RyoheiHagimoto	0:0e0631af0305	2777	matters to the compiler is the type of a and b parameters. So, if you
RyoheiHagimoto	0:0e0631af0305	2778	want a floating-point random number, but the range boundaries are
RyoheiHagimoto	0:0e0631af0305	2779	integer numbers, either put dots in the end, if they are constants, or
RyoheiHagimoto	0:0e0631af0305	2780	use explicit type cast operators, as in the a1 initialization above.
RyoheiHagimoto	0:0e0631af0305	2781	@param a lower inclusive boundary of the returned random numbers.
RyoheiHagimoto	0:0e0631af0305	2782	@param b upper non-inclusive boundary of the returned random numbers.
RyoheiHagimoto	0:0e0631af0305	2783	*/
RyoheiHagimoto	0:0e0631af0305	2784	int uniform(int a, int b);
RyoheiHagimoto	0:0e0631af0305	2785	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2786	float uniform(float a, float b);
RyoheiHagimoto	0:0e0631af0305	2787	/** @overload */
RyoheiHagimoto	0:0e0631af0305	2788	double uniform(double a, double b);
RyoheiHagimoto	0:0e0631af0305	2789
RyoheiHagimoto	0:0e0631af0305	2790	/** @brief Fills arrays with random numbers.
RyoheiHagimoto	0:0e0631af0305	2791
RyoheiHagimoto	0:0e0631af0305	2792	@param mat 2D or N-dimensional matrix; currently matrices with more than
RyoheiHagimoto	0:0e0631af0305	2793	4 channels are not supported by the methods, use Mat::reshape as a
RyoheiHagimoto	0:0e0631af0305	2794	possible workaround.
RyoheiHagimoto	0:0e0631af0305	2795	@param distType distribution type, RNG::UNIFORM or RNG::NORMAL.
RyoheiHagimoto	0:0e0631af0305	2796	@param a first distribution parameter; in case of the uniform
RyoheiHagimoto	0:0e0631af0305	2797	distribution, this is an inclusive lower boundary, in case of the normal
RyoheiHagimoto	0:0e0631af0305	2798	distribution, this is a mean value.
RyoheiHagimoto	0:0e0631af0305	2799	@param b second distribution parameter; in case of the uniform
RyoheiHagimoto	0:0e0631af0305	2800	distribution, this is a non-inclusive upper boundary, in case of the
RyoheiHagimoto	0:0e0631af0305	2801	normal distribution, this is a standard deviation (diagonal of the
RyoheiHagimoto	0:0e0631af0305	2802	standard deviation matrix or the full standard deviation matrix).
RyoheiHagimoto	0:0e0631af0305	2803	@param saturateRange pre-saturation flag; for uniform distribution only;
RyoheiHagimoto	0:0e0631af0305	2804	if true, the method will first convert a and b to the acceptable value
RyoheiHagimoto	0:0e0631af0305	2805	range (according to the mat datatype) and then will generate uniformly
RyoheiHagimoto	0:0e0631af0305	2806	distributed random numbers within the range [saturate(a), saturate(b)),
RyoheiHagimoto	0:0e0631af0305	2807	if saturateRange=false, the method will generate uniformly distributed
RyoheiHagimoto	0:0e0631af0305	2808	random numbers in the original range [a, b) and then will saturate them,
RyoheiHagimoto	0:0e0631af0305	2809	it means, for example, that
RyoheiHagimoto	0:0e0631af0305	2810	<tt>theRNG().fill(mat_8u, RNG::UNIFORM, -DBL_MAX, DBL_MAX)</tt> will likely
RyoheiHagimoto	0:0e0631af0305	2811	produce array mostly filled with 0's and 255's, since the range (0, 255)
RyoheiHagimoto	0:0e0631af0305	2812	is significantly smaller than [-DBL_MAX, DBL_MAX).
RyoheiHagimoto	0:0e0631af0305	2813
RyoheiHagimoto	0:0e0631af0305	2814	Each of the methods fills the matrix with the random values from the
RyoheiHagimoto	0:0e0631af0305	2815	specified distribution. As the new numbers are generated, the RNG state
RyoheiHagimoto	0:0e0631af0305	2816	is updated accordingly. In case of multiple-channel images, every
RyoheiHagimoto	0:0e0631af0305	2817	channel is filled independently, which means that RNG cannot generate
RyoheiHagimoto	0:0e0631af0305	2818	samples from the multi-dimensional Gaussian distribution with
RyoheiHagimoto	0:0e0631af0305	2819	non-diagonal covariance matrix directly. To do that, the method
RyoheiHagimoto	0:0e0631af0305	2820	generates samples from multi-dimensional standard Gaussian distribution
RyoheiHagimoto	0:0e0631af0305	2821	with zero mean and identity covariation matrix, and then transforms them
RyoheiHagimoto	0:0e0631af0305	2822	using transform to get samples from the specified Gaussian distribution.
RyoheiHagimoto	0:0e0631af0305	2823	*/
RyoheiHagimoto	0:0e0631af0305	2824	void fill( InputOutputArray mat, int distType, InputArray a, InputArray b, bool saturateRange = false );
RyoheiHagimoto	0:0e0631af0305	2825
RyoheiHagimoto	0:0e0631af0305	2826	/** @brief Returns the next random number sampled from the Gaussian distribution
RyoheiHagimoto	0:0e0631af0305	2827	@param sigma standard deviation of the distribution.
RyoheiHagimoto	0:0e0631af0305	2828
RyoheiHagimoto	0:0e0631af0305	2829	The method transforms the state using the MWC algorithm and returns the
RyoheiHagimoto	0:0e0631af0305	2830	next random number from the Gaussian distribution N(0,sigma) . That is,
RyoheiHagimoto	0:0e0631af0305	2831	the mean value of the returned random numbers is zero and the standard
RyoheiHagimoto	0:0e0631af0305	2832	deviation is the specified sigma .
RyoheiHagimoto	0:0e0631af0305	2833	*/
RyoheiHagimoto	0:0e0631af0305	2834	double gaussian(double sigma);
RyoheiHagimoto	0:0e0631af0305	2835
RyoheiHagimoto	0:0e0631af0305	2836	uint64 state;
RyoheiHagimoto	0:0e0631af0305	2837	};
RyoheiHagimoto	0:0e0631af0305	2838
RyoheiHagimoto	0:0e0631af0305	2839	/** @brief Mersenne Twister random number generator
RyoheiHagimoto	0:0e0631af0305	2840
RyoheiHagimoto	0:0e0631af0305	2841	Inspired by http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/MT2002/CODES/mt19937ar.c
RyoheiHagimoto	0:0e0631af0305	2842	@todo document
RyoheiHagimoto	0:0e0631af0305	2843	*/
RyoheiHagimoto	0:0e0631af0305	2844	class CV_EXPORTS RNG_MT19937
RyoheiHagimoto	0:0e0631af0305	2845	{
RyoheiHagimoto	0:0e0631af0305	2846	public:
RyoheiHagimoto	0:0e0631af0305	2847	RNG_MT19937();
RyoheiHagimoto	0:0e0631af0305	2848	RNG_MT19937(unsigned s);
RyoheiHagimoto	0:0e0631af0305	2849	void seed(unsigned s);
RyoheiHagimoto	0:0e0631af0305	2850
RyoheiHagimoto	0:0e0631af0305	2851	unsigned next();
RyoheiHagimoto	0:0e0631af0305	2852
RyoheiHagimoto	0:0e0631af0305	2853	operator int();
RyoheiHagimoto	0:0e0631af0305	2854	operator unsigned();
RyoheiHagimoto	0:0e0631af0305	2855	operator float();
RyoheiHagimoto	0:0e0631af0305	2856	operator double();
RyoheiHagimoto	0:0e0631af0305	2857
RyoheiHagimoto	0:0e0631af0305	2858	unsigned operator ()(unsigned N);
RyoheiHagimoto	0:0e0631af0305	2859	unsigned operator ()();
RyoheiHagimoto	0:0e0631af0305	2860
RyoheiHagimoto	0:0e0631af0305	2861	/** @brief returns uniformly distributed integer random number from [a,b) range
RyoheiHagimoto	0:0e0631af0305	2862
RyoheiHagimoto	0:0e0631af0305	2863	*/
RyoheiHagimoto	0:0e0631af0305	2864	int uniform(int a, int b);
RyoheiHagimoto	0:0e0631af0305	2865	/** @brief returns uniformly distributed floating-point random number from [a,b) range
RyoheiHagimoto	0:0e0631af0305	2866
RyoheiHagimoto	0:0e0631af0305	2867	*/
RyoheiHagimoto	0:0e0631af0305	2868	float uniform(float a, float b);
RyoheiHagimoto	0:0e0631af0305	2869	/** @brief returns uniformly distributed double-precision floating-point random number from [a,b) range
RyoheiHagimoto	0:0e0631af0305	2870
RyoheiHagimoto	0:0e0631af0305	2871	*/
RyoheiHagimoto	0:0e0631af0305	2872	double uniform(double a, double b);
RyoheiHagimoto	0:0e0631af0305	2873
RyoheiHagimoto	0:0e0631af0305	2874	private:
RyoheiHagimoto	0:0e0631af0305	2875	enum PeriodParameters {N = 624, M = 397};
RyoheiHagimoto	0:0e0631af0305	2876	unsigned state[N];
RyoheiHagimoto	0:0e0631af0305	2877	int mti;
RyoheiHagimoto	0:0e0631af0305	2878	};
RyoheiHagimoto	0:0e0631af0305	2879
RyoheiHagimoto	0:0e0631af0305	2880	//! @} core_array
RyoheiHagimoto	0:0e0631af0305	2881
RyoheiHagimoto	0:0e0631af0305	2882	//! @addtogroup core_cluster
RyoheiHagimoto	0:0e0631af0305	2883	//! @{
RyoheiHagimoto	0:0e0631af0305	2884
RyoheiHagimoto	0:0e0631af0305	2885	/** @example kmeans.cpp
RyoheiHagimoto	0:0e0631af0305	2886	An example on K-means clustering
RyoheiHagimoto	0:0e0631af0305	2887	*/
RyoheiHagimoto	0:0e0631af0305	2888
RyoheiHagimoto	0:0e0631af0305	2889	/** @brief Finds centers of clusters and groups input samples around the clusters.
RyoheiHagimoto	0:0e0631af0305	2890
RyoheiHagimoto	0:0e0631af0305	2891	The function kmeans implements a k-means algorithm that finds the centers of cluster_count clusters
RyoheiHagimoto	0:0e0631af0305	2892	and groups the input samples around the clusters. As an output, \f$\texttt{labels}_i\f$ contains a
RyoheiHagimoto	0:0e0631af0305	2893	0-based cluster index for the sample stored in the \f$i^{th}\f$ row of the samples matrix.
RyoheiHagimoto	0:0e0631af0305	2894
RyoheiHagimoto	0:0e0631af0305	2895	@note
RyoheiHagimoto	0:0e0631af0305	2896	- (Python) An example on K-means clustering can be found at
RyoheiHagimoto	0:0e0631af0305	2897	opencv_source_code/samples/python/kmeans.py
RyoheiHagimoto	0:0e0631af0305	2898	@param data Data for clustering. An array of N-Dimensional points with float coordinates is needed.
RyoheiHagimoto	0:0e0631af0305	2899	Examples of this array can be:
RyoheiHagimoto	0:0e0631af0305	2900	- Mat points(count, 2, CV_32F);
RyoheiHagimoto	0:0e0631af0305	2901	- Mat points(count, 1, CV_32FC2);
RyoheiHagimoto	0:0e0631af0305	2902	- Mat points(1, count, CV_32FC2);
RyoheiHagimoto	0:0e0631af0305	2903	- std::vector\<cv::Point2f\> points(sampleCount);
RyoheiHagimoto	0:0e0631af0305	2904	@param K Number of clusters to split the set by.
RyoheiHagimoto	0:0e0631af0305	2905	@param bestLabels Input/output integer array that stores the cluster indices for every sample.
RyoheiHagimoto	0:0e0631af0305	2906	@param criteria The algorithm termination criteria, that is, the maximum number of iterations and/or
RyoheiHagimoto	0:0e0631af0305	2907	the desired accuracy. The accuracy is specified as criteria.epsilon. As soon as each of the cluster
RyoheiHagimoto	0:0e0631af0305	2908	centers moves by less than criteria.epsilon on some iteration, the algorithm stops.
RyoheiHagimoto	0:0e0631af0305	2909	@param attempts Flag to specify the number of times the algorithm is executed using different
RyoheiHagimoto	0:0e0631af0305	2910	initial labellings. The algorithm returns the labels that yield the best compactness (see the last
RyoheiHagimoto	0:0e0631af0305	2911	function parameter).
RyoheiHagimoto	0:0e0631af0305	2912	@param flags Flag that can take values of cv::KmeansFlags
RyoheiHagimoto	0:0e0631af0305	2913	@param centers Output matrix of the cluster centers, one row per each cluster center.
RyoheiHagimoto	0:0e0631af0305	2914	@return The function returns the compactness measure that is computed as
RyoheiHagimoto	0:0e0631af0305	2915	\f[\sum _i \\| \texttt{samples} _i - \texttt{centers} _{ \texttt{labels} _i} \\| ^2\f]
RyoheiHagimoto	0:0e0631af0305	2916	after every attempt. The best (minimum) value is chosen and the corresponding labels and the
RyoheiHagimoto	0:0e0631af0305	2917	compactness value are returned by the function. Basically, you can use only the core of the
RyoheiHagimoto	0:0e0631af0305	2918	function, set the number of attempts to 1, initialize labels each time using a custom algorithm,
RyoheiHagimoto	0:0e0631af0305	2919	pass them with the ( flags = KMEANS_USE_INITIAL_LABELS ) flag, and then choose the best
RyoheiHagimoto	0:0e0631af0305	2920	(most-compact) clustering.
RyoheiHagimoto	0:0e0631af0305	2921	*/
RyoheiHagimoto	0:0e0631af0305	2922	CV_EXPORTS_W double kmeans( InputArray data, int K, InputOutputArray bestLabels,
RyoheiHagimoto	0:0e0631af0305	2923	TermCriteria criteria, int attempts,
RyoheiHagimoto	0:0e0631af0305	2924	int flags, OutputArray centers = noArray() );
RyoheiHagimoto	0:0e0631af0305	2925
RyoheiHagimoto	0:0e0631af0305	2926	//! @} core_cluster
RyoheiHagimoto	0:0e0631af0305	2927
RyoheiHagimoto	0:0e0631af0305	2928	//! @addtogroup core_basic
RyoheiHagimoto	0:0e0631af0305	2929	//! @{
RyoheiHagimoto	0:0e0631af0305	2930
RyoheiHagimoto	0:0e0631af0305	2931	/////////////////////////////// Formatted output of cv::Mat ///////////////////////////
RyoheiHagimoto	0:0e0631af0305	2932
RyoheiHagimoto	0:0e0631af0305	2933	/** @todo document */
RyoheiHagimoto	0:0e0631af0305	2934	class CV_EXPORTS Formatted
RyoheiHagimoto	0:0e0631af0305	2935	{
RyoheiHagimoto	0:0e0631af0305	2936	public:
RyoheiHagimoto	0:0e0631af0305	2937	virtual const char* next() = 0;
RyoheiHagimoto	0:0e0631af0305	2938	virtual void reset() = 0;
RyoheiHagimoto	0:0e0631af0305	2939	virtual ~Formatted();
RyoheiHagimoto	0:0e0631af0305	2940	};
RyoheiHagimoto	0:0e0631af0305	2941
RyoheiHagimoto	0:0e0631af0305	2942	/** @todo document */
RyoheiHagimoto	0:0e0631af0305	2943	class CV_EXPORTS Formatter
RyoheiHagimoto	0:0e0631af0305	2944	{
RyoheiHagimoto	0:0e0631af0305	2945	public:
RyoheiHagimoto	0:0e0631af0305	2946	enum { FMT_DEFAULT = 0,
RyoheiHagimoto	0:0e0631af0305	2947	FMT_MATLAB = 1,
RyoheiHagimoto	0:0e0631af0305	2948	FMT_CSV = 2,
RyoheiHagimoto	0:0e0631af0305	2949	FMT_PYTHON = 3,
RyoheiHagimoto	0:0e0631af0305	2950	FMT_NUMPY = 4,
RyoheiHagimoto	0:0e0631af0305	2951	FMT_C = 5
RyoheiHagimoto	0:0e0631af0305	2952	};
RyoheiHagimoto	0:0e0631af0305	2953
RyoheiHagimoto	0:0e0631af0305	2954	virtual ~Formatter();
RyoheiHagimoto	0:0e0631af0305	2955
RyoheiHagimoto	0:0e0631af0305	2956	virtual Ptr<Formatted> format(const Mat& mtx) const = 0;
RyoheiHagimoto	0:0e0631af0305	2957
RyoheiHagimoto	0:0e0631af0305	2958	virtual void set32fPrecision(int p = 8) = 0;
RyoheiHagimoto	0:0e0631af0305	2959	virtual void set64fPrecision(int p = 16) = 0;
RyoheiHagimoto	0:0e0631af0305	2960	virtual void setMultiline(bool ml = true) = 0;
RyoheiHagimoto	0:0e0631af0305	2961
RyoheiHagimoto	0:0e0631af0305	2962	static Ptr<Formatter> get(int fmt = FMT_DEFAULT);
RyoheiHagimoto	0:0e0631af0305	2963
RyoheiHagimoto	0:0e0631af0305	2964	};
RyoheiHagimoto	0:0e0631af0305	2965
RyoheiHagimoto	0:0e0631af0305	2966	static inline
RyoheiHagimoto	0:0e0631af0305	2967	String& operator << (String& out, Ptr<Formatted> fmtd)
RyoheiHagimoto	0:0e0631af0305	2968	{
RyoheiHagimoto	0:0e0631af0305	2969	fmtd->reset();
RyoheiHagimoto	0:0e0631af0305	2970	for(const char* str = fmtd->next(); str; str = fmtd->next())
RyoheiHagimoto	0:0e0631af0305	2971	out += cv::String(str);
RyoheiHagimoto	0:0e0631af0305	2972	return out;
RyoheiHagimoto	0:0e0631af0305	2973	}
RyoheiHagimoto	0:0e0631af0305	2974
RyoheiHagimoto	0:0e0631af0305	2975	static inline
RyoheiHagimoto	0:0e0631af0305	2976	String& operator << (String& out, const Mat& mtx)
RyoheiHagimoto	0:0e0631af0305	2977	{
RyoheiHagimoto	0:0e0631af0305	2978	return out << Formatter::get()->format(mtx);
RyoheiHagimoto	0:0e0631af0305	2979	}
RyoheiHagimoto	0:0e0631af0305	2980
RyoheiHagimoto	0:0e0631af0305	2981	//////////////////////////////////////// Algorithm ////////////////////////////////////
RyoheiHagimoto	0:0e0631af0305	2982
RyoheiHagimoto	0:0e0631af0305	2983	class CV_EXPORTS Algorithm;
RyoheiHagimoto	0:0e0631af0305	2984
RyoheiHagimoto	0:0e0631af0305	2985	template<typename _Tp> struct ParamType {};
RyoheiHagimoto	0:0e0631af0305	2986
RyoheiHagimoto	0:0e0631af0305	2987
RyoheiHagimoto	0:0e0631af0305	2988	/** @brief This is a base class for all more or less complex algorithms in OpenCV
RyoheiHagimoto	0:0e0631af0305	2989
RyoheiHagimoto	0:0e0631af0305	2990	especially for classes of algorithms, for which there can be multiple implementations. The examples
RyoheiHagimoto	0:0e0631af0305	2991	are stereo correspondence (for which there are algorithms like block matching, semi-global block
RyoheiHagimoto	0:0e0631af0305	2992	matching, graph-cut etc.), background subtraction (which can be done using mixture-of-gaussians
RyoheiHagimoto	0:0e0631af0305	2993	models, codebook-based algorithm etc.), optical flow (block matching, Lucas-Kanade, Horn-Schunck
RyoheiHagimoto	0:0e0631af0305	2994	etc.).
RyoheiHagimoto	0:0e0631af0305	2995
RyoheiHagimoto	0:0e0631af0305	2996	Here is example of SIFT use in your application via Algorithm interface:
RyoheiHagimoto	0:0e0631af0305	2997	@code
RyoheiHagimoto	0:0e0631af0305	2998	#include "opencv2/opencv.hpp"
RyoheiHagimoto	0:0e0631af0305	2999	#include "opencv2/xfeatures2d.hpp"
RyoheiHagimoto	0:0e0631af0305	3000	using namespace cv::xfeatures2d;
RyoheiHagimoto	0:0e0631af0305	3001
RyoheiHagimoto	0:0e0631af0305	3002	Ptr<Feature2D> sift = SIFT::create();
RyoheiHagimoto	0:0e0631af0305	3003	FileStorage fs("sift_params.xml", FileStorage::READ);
RyoheiHagimoto	0:0e0631af0305	3004	if( fs.isOpened() ) // if we have file with parameters, read them
RyoheiHagimoto	0:0e0631af0305	3005	{
RyoheiHagimoto	0:0e0631af0305	3006	sift->read(fs["sift_params"]);
RyoheiHagimoto	0:0e0631af0305	3007	fs.release();
RyoheiHagimoto	0:0e0631af0305	3008	}
RyoheiHagimoto	0:0e0631af0305	3009	else // else modify the parameters and store them; user can later edit the file to use different parameters
RyoheiHagimoto	0:0e0631af0305	3010	{
RyoheiHagimoto	0:0e0631af0305	3011	sift->setContrastThreshold(0.01f); // lower the contrast threshold, compared to the default value
RyoheiHagimoto	0:0e0631af0305	3012	{
RyoheiHagimoto	0:0e0631af0305	3013	WriteStructContext ws(fs, "sift_params", CV_NODE_MAP);
RyoheiHagimoto	0:0e0631af0305	3014	sift->write(fs);
RyoheiHagimoto	0:0e0631af0305	3015	}
RyoheiHagimoto	0:0e0631af0305	3016	}
RyoheiHagimoto	0:0e0631af0305	3017	Mat image = imread("myimage.png", 0), descriptors;
RyoheiHagimoto	0:0e0631af0305	3018	vector<KeyPoint> keypoints;
RyoheiHagimoto	0:0e0631af0305	3019	sift->detectAndCompute(image, noArray(), keypoints, descriptors);
RyoheiHagimoto	0:0e0631af0305	3020	@endcode
RyoheiHagimoto	0:0e0631af0305	3021	*/
RyoheiHagimoto	0:0e0631af0305	3022	class CV_EXPORTS_W Algorithm
RyoheiHagimoto	0:0e0631af0305	3023	{
RyoheiHagimoto	0:0e0631af0305	3024	public:
RyoheiHagimoto	0:0e0631af0305	3025	Algorithm();
RyoheiHagimoto	0:0e0631af0305	3026	virtual ~Algorithm();
RyoheiHagimoto	0:0e0631af0305	3027
RyoheiHagimoto	0:0e0631af0305	3028	/** @brief Clears the algorithm state
RyoheiHagimoto	0:0e0631af0305	3029	*/
RyoheiHagimoto	0:0e0631af0305	3030	CV_WRAP virtual void clear() {}
RyoheiHagimoto	0:0e0631af0305	3031
RyoheiHagimoto	0:0e0631af0305	3032	/** @brief Stores algorithm parameters in a file storage
RyoheiHagimoto	0:0e0631af0305	3033	*/
RyoheiHagimoto	0:0e0631af0305	3034	virtual void write(FileStorage& fs) const { (void)fs; }
RyoheiHagimoto	0:0e0631af0305	3035
RyoheiHagimoto	0:0e0631af0305	3036	/** @brief Reads algorithm parameters from a file storage
RyoheiHagimoto	0:0e0631af0305	3037	*/
RyoheiHagimoto	0:0e0631af0305	3038	virtual void read(const FileNode& fn) { (void)fn; }
RyoheiHagimoto	0:0e0631af0305	3039
RyoheiHagimoto	0:0e0631af0305	3040	/** @brief Returns true if the Algorithm is empty (e.g. in the very beginning or after unsuccessful read
RyoheiHagimoto	0:0e0631af0305	3041	*/
RyoheiHagimoto	0:0e0631af0305	3042	virtual bool empty() const { return false; }
RyoheiHagimoto	0:0e0631af0305	3043
RyoheiHagimoto	0:0e0631af0305	3044	/** @brief Reads algorithm from the file node
RyoheiHagimoto	0:0e0631af0305	3045
RyoheiHagimoto	0:0e0631af0305	3046	This is static template method of Algorithm. It's usage is following (in the case of SVM):
RyoheiHagimoto	0:0e0631af0305	3047	@code
RyoheiHagimoto	0:0e0631af0305	3048	cv::FileStorage fsRead("example.xml", FileStorage::READ);
RyoheiHagimoto	0:0e0631af0305	3049	Ptr<SVM> svm = Algorithm::read<SVM>(fsRead.root());
RyoheiHagimoto	0:0e0631af0305	3050	@endcode
RyoheiHagimoto	0:0e0631af0305	3051	In order to make this method work, the derived class must overwrite Algorithm::read(const
RyoheiHagimoto	0:0e0631af0305	3052	FileNode& fn) and also have static create() method without parameters
RyoheiHagimoto	0:0e0631af0305	3053	(or with all the optional parameters)
RyoheiHagimoto	0:0e0631af0305	3054	*/
RyoheiHagimoto	0:0e0631af0305	3055	template<typename _Tp> static Ptr<_Tp> read(const FileNode& fn)
RyoheiHagimoto	0:0e0631af0305	3056	{
RyoheiHagimoto	0:0e0631af0305	3057	Ptr<_Tp> obj = _Tp::create();
RyoheiHagimoto	0:0e0631af0305	3058	obj->read(fn);
RyoheiHagimoto	0:0e0631af0305	3059	return !obj->empty() ? obj : Ptr<_Tp>();
RyoheiHagimoto	0:0e0631af0305	3060	}
RyoheiHagimoto	0:0e0631af0305	3061
RyoheiHagimoto	0:0e0631af0305	3062	/** @brief Loads algorithm from the file
RyoheiHagimoto	0:0e0631af0305	3063
RyoheiHagimoto	0:0e0631af0305	3064	@param filename Name of the file to read.
RyoheiHagimoto	0:0e0631af0305	3065	@param objname The optional name of the node to read (if empty, the first top-level node will be used)
RyoheiHagimoto	0:0e0631af0305	3066
RyoheiHagimoto	0:0e0631af0305	3067	This is static template method of Algorithm. It's usage is following (in the case of SVM):
RyoheiHagimoto	0:0e0631af0305	3068	@code
RyoheiHagimoto	0:0e0631af0305	3069	Ptr<SVM> svm = Algorithm::load<SVM>("my_svm_model.xml");
RyoheiHagimoto	0:0e0631af0305	3070	@endcode
RyoheiHagimoto	0:0e0631af0305	3071	In order to make this method work, the derived class must overwrite Algorithm::read(const
RyoheiHagimoto	0:0e0631af0305	3072	FileNode& fn).
RyoheiHagimoto	0:0e0631af0305	3073	*/
RyoheiHagimoto	0:0e0631af0305	3074	template<typename _Tp> static Ptr<_Tp> load(const String& filename, const String& objname=String())
RyoheiHagimoto	0:0e0631af0305	3075	{
RyoheiHagimoto	0:0e0631af0305	3076	FileStorage fs(filename, FileStorage::READ);
RyoheiHagimoto	0:0e0631af0305	3077	FileNode fn = objname.empty() ? fs.getFirstTopLevelNode() : fs[objname];
RyoheiHagimoto	0:0e0631af0305	3078	if (fn.empty()) return Ptr<_Tp>();
RyoheiHagimoto	0:0e0631af0305	3079	Ptr<_Tp> obj = _Tp::create();
RyoheiHagimoto	0:0e0631af0305	3080	obj->read(fn);
RyoheiHagimoto	0:0e0631af0305	3081	return !obj->empty() ? obj : Ptr<_Tp>();
RyoheiHagimoto	0:0e0631af0305	3082	}
RyoheiHagimoto	0:0e0631af0305	3083
RyoheiHagimoto	0:0e0631af0305	3084	/** @brief Loads algorithm from a String
RyoheiHagimoto	0:0e0631af0305	3085
RyoheiHagimoto	0:0e0631af0305	3086	@param strModel The string variable containing the model you want to load.
RyoheiHagimoto	0:0e0631af0305	3087	@param objname The optional name of the node to read (if empty, the first top-level node will be used)
RyoheiHagimoto	0:0e0631af0305	3088
RyoheiHagimoto	0:0e0631af0305	3089	This is static template method of Algorithm. It's usage is following (in the case of SVM):
RyoheiHagimoto	0:0e0631af0305	3090	@code
RyoheiHagimoto	0:0e0631af0305	3091	Ptr<SVM> svm = Algorithm::loadFromString<SVM>(myStringModel);
RyoheiHagimoto	0:0e0631af0305	3092	@endcode
RyoheiHagimoto	0:0e0631af0305	3093	*/
RyoheiHagimoto	0:0e0631af0305	3094	template<typename _Tp> static Ptr<_Tp> loadFromString(const String& strModel, const String& objname=String())
RyoheiHagimoto	0:0e0631af0305	3095	{
RyoheiHagimoto	0:0e0631af0305	3096	FileStorage fs(strModel, FileStorage::READ + FileStorage::MEMORY);
RyoheiHagimoto	0:0e0631af0305	3097	FileNode fn = objname.empty() ? fs.getFirstTopLevelNode() : fs[objname];
RyoheiHagimoto	0:0e0631af0305	3098	Ptr<_Tp> obj = _Tp::create();
RyoheiHagimoto	0:0e0631af0305	3099	obj->read(fn);
RyoheiHagimoto	0:0e0631af0305	3100	return !obj->empty() ? obj : Ptr<_Tp>();
RyoheiHagimoto	0:0e0631af0305	3101	}
RyoheiHagimoto	0:0e0631af0305	3102
RyoheiHagimoto	0:0e0631af0305	3103	/** Saves the algorithm to a file.
RyoheiHagimoto	0:0e0631af0305	3104	In order to make this method work, the derived class must implement Algorithm::write(FileStorage& fs). */
RyoheiHagimoto	0:0e0631af0305	3105	CV_WRAP virtual void save(const String& filename) const;
RyoheiHagimoto	0:0e0631af0305	3106
RyoheiHagimoto	0:0e0631af0305	3107	/** Returns the algorithm string identifier.
RyoheiHagimoto	0:0e0631af0305	3108	This string is used as top level xml/yml node tag when the object is saved to a file or string. */
RyoheiHagimoto	0:0e0631af0305	3109	CV_WRAP virtual String getDefaultName() const;
RyoheiHagimoto	0:0e0631af0305	3110
RyoheiHagimoto	0:0e0631af0305	3111	protected:
RyoheiHagimoto	0:0e0631af0305	3112	void writeFormat(FileStorage& fs) const;
RyoheiHagimoto	0:0e0631af0305	3113	};
RyoheiHagimoto	0:0e0631af0305	3114
RyoheiHagimoto	0:0e0631af0305	3115	struct Param {
RyoheiHagimoto	0:0e0631af0305	3116	enum { INT=0, BOOLEAN=1, REAL=2, STRING=3, MAT=4, MAT_VECTOR=5, ALGORITHM=6, FLOAT=7,
RyoheiHagimoto	0:0e0631af0305	3117	UNSIGNED_INT=8, UINT64=9, UCHAR=11 };
RyoheiHagimoto	0:0e0631af0305	3118	};
RyoheiHagimoto	0:0e0631af0305	3119
RyoheiHagimoto	0:0e0631af0305	3120
RyoheiHagimoto	0:0e0631af0305	3121
RyoheiHagimoto	0:0e0631af0305	3122	template<> struct ParamType<bool>
RyoheiHagimoto	0:0e0631af0305	3123	{
RyoheiHagimoto	0:0e0631af0305	3124	typedef bool const_param_type;
RyoheiHagimoto	0:0e0631af0305	3125	typedef bool member_type;
RyoheiHagimoto	0:0e0631af0305	3126
RyoheiHagimoto	0:0e0631af0305	3127	enum { type = Param::BOOLEAN };
RyoheiHagimoto	0:0e0631af0305	3128	};
RyoheiHagimoto	0:0e0631af0305	3129
RyoheiHagimoto	0:0e0631af0305	3130	template<> struct ParamType<int>
RyoheiHagimoto	0:0e0631af0305	3131	{
RyoheiHagimoto	0:0e0631af0305	3132	typedef int const_param_type;
RyoheiHagimoto	0:0e0631af0305	3133	typedef int member_type;
RyoheiHagimoto	0:0e0631af0305	3134
RyoheiHagimoto	0:0e0631af0305	3135	enum { type = Param::INT };
RyoheiHagimoto	0:0e0631af0305	3136	};
RyoheiHagimoto	0:0e0631af0305	3137
RyoheiHagimoto	0:0e0631af0305	3138	template<> struct ParamType<double>
RyoheiHagimoto	0:0e0631af0305	3139	{
RyoheiHagimoto	0:0e0631af0305	3140	typedef double const_param_type;
RyoheiHagimoto	0:0e0631af0305	3141	typedef double member_type;
RyoheiHagimoto	0:0e0631af0305	3142
RyoheiHagimoto	0:0e0631af0305	3143	enum { type = Param::REAL };
RyoheiHagimoto	0:0e0631af0305	3144	};
RyoheiHagimoto	0:0e0631af0305	3145
RyoheiHagimoto	0:0e0631af0305	3146	template<> struct ParamType<String>
RyoheiHagimoto	0:0e0631af0305	3147	{
RyoheiHagimoto	0:0e0631af0305	3148	typedef const String& const_param_type;
RyoheiHagimoto	0:0e0631af0305	3149	typedef String member_type;
RyoheiHagimoto	0:0e0631af0305	3150
RyoheiHagimoto	0:0e0631af0305	3151	enum { type = Param::STRING };
RyoheiHagimoto	0:0e0631af0305	3152	};
RyoheiHagimoto	0:0e0631af0305	3153
RyoheiHagimoto	0:0e0631af0305	3154	template<> struct ParamType<Mat>
RyoheiHagimoto	0:0e0631af0305	3155	{
RyoheiHagimoto	0:0e0631af0305	3156	typedef const Mat& const_param_type;
RyoheiHagimoto	0:0e0631af0305	3157	typedef Mat member_type;
RyoheiHagimoto	0:0e0631af0305	3158
RyoheiHagimoto	0:0e0631af0305	3159	enum { type = Param::MAT };
RyoheiHagimoto	0:0e0631af0305	3160	};
RyoheiHagimoto	0:0e0631af0305	3161
RyoheiHagimoto	0:0e0631af0305	3162	template<> struct ParamType<std::vector<Mat> >
RyoheiHagimoto	0:0e0631af0305	3163	{
RyoheiHagimoto	0:0e0631af0305	3164	typedef const std::vector<Mat>& const_param_type;
RyoheiHagimoto	0:0e0631af0305	3165	typedef std::vector<Mat> member_type;
RyoheiHagimoto	0:0e0631af0305	3166
RyoheiHagimoto	0:0e0631af0305	3167	enum { type = Param::MAT_VECTOR };
RyoheiHagimoto	0:0e0631af0305	3168	};
RyoheiHagimoto	0:0e0631af0305	3169
RyoheiHagimoto	0:0e0631af0305	3170	template<> struct ParamType<Algorithm>
RyoheiHagimoto	0:0e0631af0305	3171	{
RyoheiHagimoto	0:0e0631af0305	3172	typedef const Ptr<Algorithm>& const_param_type;
RyoheiHagimoto	0:0e0631af0305	3173	typedef Ptr<Algorithm> member_type;
RyoheiHagimoto	0:0e0631af0305	3174
RyoheiHagimoto	0:0e0631af0305	3175	enum { type = Param::ALGORITHM };
RyoheiHagimoto	0:0e0631af0305	3176	};
RyoheiHagimoto	0:0e0631af0305	3177
RyoheiHagimoto	0:0e0631af0305	3178	template<> struct ParamType<float>
RyoheiHagimoto	0:0e0631af0305	3179	{
RyoheiHagimoto	0:0e0631af0305	3180	typedef float const_param_type;
RyoheiHagimoto	0:0e0631af0305	3181	typedef float member_type;
RyoheiHagimoto	0:0e0631af0305	3182
RyoheiHagimoto	0:0e0631af0305	3183	enum { type = Param::FLOAT };
RyoheiHagimoto	0:0e0631af0305	3184	};
RyoheiHagimoto	0:0e0631af0305	3185
RyoheiHagimoto	0:0e0631af0305	3186	template<> struct ParamType<unsigned>
RyoheiHagimoto	0:0e0631af0305	3187	{
RyoheiHagimoto	0:0e0631af0305	3188	typedef unsigned const_param_type;
RyoheiHagimoto	0:0e0631af0305	3189	typedef unsigned member_type;
RyoheiHagimoto	0:0e0631af0305	3190
RyoheiHagimoto	0:0e0631af0305	3191	enum { type = Param::UNSIGNED_INT };
RyoheiHagimoto	0:0e0631af0305	3192	};
RyoheiHagimoto	0:0e0631af0305	3193
RyoheiHagimoto	0:0e0631af0305	3194	template<> struct ParamType<uint64>
RyoheiHagimoto	0:0e0631af0305	3195	{
RyoheiHagimoto	0:0e0631af0305	3196	typedef uint64 const_param_type;
RyoheiHagimoto	0:0e0631af0305	3197	typedef uint64 member_type;
RyoheiHagimoto	0:0e0631af0305	3198
RyoheiHagimoto	0:0e0631af0305	3199	enum { type = Param::UINT64 };
RyoheiHagimoto	0:0e0631af0305	3200	};
RyoheiHagimoto	0:0e0631af0305	3201
RyoheiHagimoto	0:0e0631af0305	3202	template<> struct ParamType<uchar>
RyoheiHagimoto	0:0e0631af0305	3203	{
RyoheiHagimoto	0:0e0631af0305	3204	typedef uchar const_param_type;
RyoheiHagimoto	0:0e0631af0305	3205	typedef uchar member_type;
RyoheiHagimoto	0:0e0631af0305	3206
RyoheiHagimoto	0:0e0631af0305	3207	enum { type = Param::UCHAR };
RyoheiHagimoto	0:0e0631af0305	3208	};
RyoheiHagimoto	0:0e0631af0305	3209
RyoheiHagimoto	0:0e0631af0305	3210	//! @} core_basic
RyoheiHagimoto	0:0e0631af0305	3211
RyoheiHagimoto	0:0e0631af0305	3212	} //namespace cv
RyoheiHagimoto	0:0e0631af0305	3213
RyoheiHagimoto	0:0e0631af0305	3214	#include "opencv2/core/operations.hpp"
RyoheiHagimoto	0:0e0631af0305	3215	#include "opencv2/core/cvstd.inl.hpp"
RyoheiHagimoto	0:0e0631af0305	3216	#include "opencv2/core/utility.hpp"
RyoheiHagimoto	0:0e0631af0305	3217	#include "opencv2/core/optim.hpp"
RyoheiHagimoto	0:0e0631af0305	3218	#include "opencv2/core/ovx.hpp"
RyoheiHagimoto	0:0e0631af0305	3219
RyoheiHagimoto	0:0e0631af0305	3220	#endif /OPENCV_CORE_HPP/

Repository toolbox

Export to desktop IDE

Repository details

Type:	Library
Created:	29 Jan 2021
Imports:	11
Forks:	0
Commits:	1
Dependents:	1
Dependencies:	0
Followers:	60

include/opencv2/core.hpp@0:0e0631af0305, 2021-01-29 (annotated)

Who changed what in which revision?

Repository toolbox

Repository details

Important Information for this Arm website

Access Warning