Renesas / opencv-lib

openCV library for Renesas RZ/A

Dependents:   RZ_A2M_Mbed_samples

include/opencv2/video/tracking.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?

UserRevisionLine numberNew 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) 2000-2008, Intel Corporation, all rights reserved.
RyoheiHagimoto 0:0e0631af0305 14 // Copyright (C) 2009, Willow Garage Inc., all rights reserved.
RyoheiHagimoto 0:0e0631af0305 15 // Copyright (C) 2013, OpenCV Foundation, all rights reserved.
RyoheiHagimoto 0:0e0631af0305 16 // Third party copyrights are property of their respective owners.
RyoheiHagimoto 0:0e0631af0305 17 //
RyoheiHagimoto 0:0e0631af0305 18 // Redistribution and use in source and binary forms, with or without modification,
RyoheiHagimoto 0:0e0631af0305 19 // are permitted provided that the following conditions are met:
RyoheiHagimoto 0:0e0631af0305 20 //
RyoheiHagimoto 0:0e0631af0305 21 // * Redistribution's of source code must retain the above copyright notice,
RyoheiHagimoto 0:0e0631af0305 22 // this list of conditions and the following disclaimer.
RyoheiHagimoto 0:0e0631af0305 23 //
RyoheiHagimoto 0:0e0631af0305 24 // * Redistribution's in binary form must reproduce the above copyright notice,
RyoheiHagimoto 0:0e0631af0305 25 // this list of conditions and the following disclaimer in the documentation
RyoheiHagimoto 0:0e0631af0305 26 // and/or other materials provided with the distribution.
RyoheiHagimoto 0:0e0631af0305 27 //
RyoheiHagimoto 0:0e0631af0305 28 // * The name of the copyright holders may not be used to endorse or promote products
RyoheiHagimoto 0:0e0631af0305 29 // derived from this software without specific prior written permission.
RyoheiHagimoto 0:0e0631af0305 30 //
RyoheiHagimoto 0:0e0631af0305 31 // This software is provided by the copyright holders and contributors "as is" and
RyoheiHagimoto 0:0e0631af0305 32 // any express or implied warranties, including, but not limited to, the implied
RyoheiHagimoto 0:0e0631af0305 33 // warranties of merchantability and fitness for a particular purpose are disclaimed.
RyoheiHagimoto 0:0e0631af0305 34 // In no event shall the Intel Corporation or contributors be liable for any direct,
RyoheiHagimoto 0:0e0631af0305 35 // indirect, incidental, special, exemplary, or consequential damages
RyoheiHagimoto 0:0e0631af0305 36 // (including, but not limited to, procurement of substitute goods or services;
RyoheiHagimoto 0:0e0631af0305 37 // loss of use, data, or profits; or business interruption) however caused
RyoheiHagimoto 0:0e0631af0305 38 // and on any theory of liability, whether in contract, strict liability,
RyoheiHagimoto 0:0e0631af0305 39 // or tort (including negligence or otherwise) arising in any way out of
RyoheiHagimoto 0:0e0631af0305 40 // the use of this software, even if advised of the possibility of such damage.
RyoheiHagimoto 0:0e0631af0305 41 //
RyoheiHagimoto 0:0e0631af0305 42 //M*/
RyoheiHagimoto 0:0e0631af0305 43
RyoheiHagimoto 0:0e0631af0305 44 #ifndef OPENCV_TRACKING_HPP
RyoheiHagimoto 0:0e0631af0305 45 #define OPENCV_TRACKING_HPP
RyoheiHagimoto 0:0e0631af0305 46
RyoheiHagimoto 0:0e0631af0305 47 #include "opencv2/core.hpp"
RyoheiHagimoto 0:0e0631af0305 48 #include "opencv2/imgproc.hpp"
RyoheiHagimoto 0:0e0631af0305 49
RyoheiHagimoto 0:0e0631af0305 50 namespace cv
RyoheiHagimoto 0:0e0631af0305 51 {
RyoheiHagimoto 0:0e0631af0305 52
RyoheiHagimoto 0:0e0631af0305 53 //! @addtogroup video_track
RyoheiHagimoto 0:0e0631af0305 54 //! @{
RyoheiHagimoto 0:0e0631af0305 55
RyoheiHagimoto 0:0e0631af0305 56 enum { OPTFLOW_USE_INITIAL_FLOW = 4,
RyoheiHagimoto 0:0e0631af0305 57 OPTFLOW_LK_GET_MIN_EIGENVALS = 8,
RyoheiHagimoto 0:0e0631af0305 58 OPTFLOW_FARNEBACK_GAUSSIAN = 256
RyoheiHagimoto 0:0e0631af0305 59 };
RyoheiHagimoto 0:0e0631af0305 60
RyoheiHagimoto 0:0e0631af0305 61 /** @brief Finds an object center, size, and orientation.
RyoheiHagimoto 0:0e0631af0305 62
RyoheiHagimoto 0:0e0631af0305 63 @param probImage Back projection of the object histogram. See calcBackProject.
RyoheiHagimoto 0:0e0631af0305 64 @param window Initial search window.
RyoheiHagimoto 0:0e0631af0305 65 @param criteria Stop criteria for the underlying meanShift.
RyoheiHagimoto 0:0e0631af0305 66 returns
RyoheiHagimoto 0:0e0631af0305 67 (in old interfaces) Number of iterations CAMSHIFT took to converge
RyoheiHagimoto 0:0e0631af0305 68 The function implements the CAMSHIFT object tracking algorithm @cite Bradski98 . First, it finds an
RyoheiHagimoto 0:0e0631af0305 69 object center using meanShift and then adjusts the window size and finds the optimal rotation. The
RyoheiHagimoto 0:0e0631af0305 70 function returns the rotated rectangle structure that includes the object position, size, and
RyoheiHagimoto 0:0e0631af0305 71 orientation. The next position of the search window can be obtained with RotatedRect::boundingRect()
RyoheiHagimoto 0:0e0631af0305 72
RyoheiHagimoto 0:0e0631af0305 73 See the OpenCV sample camshiftdemo.c that tracks colored objects.
RyoheiHagimoto 0:0e0631af0305 74
RyoheiHagimoto 0:0e0631af0305 75 @note
RyoheiHagimoto 0:0e0631af0305 76 - (Python) A sample explaining the camshift tracking algorithm can be found at
RyoheiHagimoto 0:0e0631af0305 77 opencv_source_code/samples/python/camshift.py
RyoheiHagimoto 0:0e0631af0305 78 */
RyoheiHagimoto 0:0e0631af0305 79 CV_EXPORTS_W RotatedRect CamShift( InputArray probImage, CV_IN_OUT Rect& window,
RyoheiHagimoto 0:0e0631af0305 80 TermCriteria criteria );
RyoheiHagimoto 0:0e0631af0305 81
RyoheiHagimoto 0:0e0631af0305 82 /** @brief Finds an object on a back projection image.
RyoheiHagimoto 0:0e0631af0305 83
RyoheiHagimoto 0:0e0631af0305 84 @param probImage Back projection of the object histogram. See calcBackProject for details.
RyoheiHagimoto 0:0e0631af0305 85 @param window Initial search window.
RyoheiHagimoto 0:0e0631af0305 86 @param criteria Stop criteria for the iterative search algorithm.
RyoheiHagimoto 0:0e0631af0305 87 returns
RyoheiHagimoto 0:0e0631af0305 88 : Number of iterations CAMSHIFT took to converge.
RyoheiHagimoto 0:0e0631af0305 89 The function implements the iterative object search algorithm. It takes the input back projection of
RyoheiHagimoto 0:0e0631af0305 90 an object and the initial position. The mass center in window of the back projection image is
RyoheiHagimoto 0:0e0631af0305 91 computed and the search window center shifts to the mass center. The procedure is repeated until the
RyoheiHagimoto 0:0e0631af0305 92 specified number of iterations criteria.maxCount is done or until the window center shifts by less
RyoheiHagimoto 0:0e0631af0305 93 than criteria.epsilon. The algorithm is used inside CamShift and, unlike CamShift , the search
RyoheiHagimoto 0:0e0631af0305 94 window size or orientation do not change during the search. You can simply pass the output of
RyoheiHagimoto 0:0e0631af0305 95 calcBackProject to this function. But better results can be obtained if you pre-filter the back
RyoheiHagimoto 0:0e0631af0305 96 projection and remove the noise. For example, you can do this by retrieving connected components
RyoheiHagimoto 0:0e0631af0305 97 with findContours , throwing away contours with small area ( contourArea ), and rendering the
RyoheiHagimoto 0:0e0631af0305 98 remaining contours with drawContours.
RyoheiHagimoto 0:0e0631af0305 99
RyoheiHagimoto 0:0e0631af0305 100 @note
RyoheiHagimoto 0:0e0631af0305 101 - A mean-shift tracking sample can be found at opencv_source_code/samples/cpp/camshiftdemo.cpp
RyoheiHagimoto 0:0e0631af0305 102 */
RyoheiHagimoto 0:0e0631af0305 103 CV_EXPORTS_W int meanShift( InputArray probImage, CV_IN_OUT Rect& window, TermCriteria criteria );
RyoheiHagimoto 0:0e0631af0305 104
RyoheiHagimoto 0:0e0631af0305 105 /** @brief Constructs the image pyramid which can be passed to calcOpticalFlowPyrLK.
RyoheiHagimoto 0:0e0631af0305 106
RyoheiHagimoto 0:0e0631af0305 107 @param img 8-bit input image.
RyoheiHagimoto 0:0e0631af0305 108 @param pyramid output pyramid.
RyoheiHagimoto 0:0e0631af0305 109 @param winSize window size of optical flow algorithm. Must be not less than winSize argument of
RyoheiHagimoto 0:0e0631af0305 110 calcOpticalFlowPyrLK. It is needed to calculate required padding for pyramid levels.
RyoheiHagimoto 0:0e0631af0305 111 @param maxLevel 0-based maximal pyramid level number.
RyoheiHagimoto 0:0e0631af0305 112 @param withDerivatives set to precompute gradients for the every pyramid level. If pyramid is
RyoheiHagimoto 0:0e0631af0305 113 constructed without the gradients then calcOpticalFlowPyrLK will calculate them internally.
RyoheiHagimoto 0:0e0631af0305 114 @param pyrBorder the border mode for pyramid layers.
RyoheiHagimoto 0:0e0631af0305 115 @param derivBorder the border mode for gradients.
RyoheiHagimoto 0:0e0631af0305 116 @param tryReuseInputImage put ROI of input image into the pyramid if possible. You can pass false
RyoheiHagimoto 0:0e0631af0305 117 to force data copying.
RyoheiHagimoto 0:0e0631af0305 118 @return number of levels in constructed pyramid. Can be less than maxLevel.
RyoheiHagimoto 0:0e0631af0305 119 */
RyoheiHagimoto 0:0e0631af0305 120 CV_EXPORTS_W int buildOpticalFlowPyramid( InputArray img, OutputArrayOfArrays pyramid,
RyoheiHagimoto 0:0e0631af0305 121 Size winSize, int maxLevel, bool withDerivatives = true,
RyoheiHagimoto 0:0e0631af0305 122 int pyrBorder = BORDER_REFLECT_101,
RyoheiHagimoto 0:0e0631af0305 123 int derivBorder = BORDER_CONSTANT,
RyoheiHagimoto 0:0e0631af0305 124 bool tryReuseInputImage = true );
RyoheiHagimoto 0:0e0631af0305 125
RyoheiHagimoto 0:0e0631af0305 126 /** @brief Calculates an optical flow for a sparse feature set using the iterative Lucas-Kanade method with
RyoheiHagimoto 0:0e0631af0305 127 pyramids.
RyoheiHagimoto 0:0e0631af0305 128
RyoheiHagimoto 0:0e0631af0305 129 @param prevImg first 8-bit input image or pyramid constructed by buildOpticalFlowPyramid.
RyoheiHagimoto 0:0e0631af0305 130 @param nextImg second input image or pyramid of the same size and the same type as prevImg.
RyoheiHagimoto 0:0e0631af0305 131 @param prevPts vector of 2D points for which the flow needs to be found; point coordinates must be
RyoheiHagimoto 0:0e0631af0305 132 single-precision floating-point numbers.
RyoheiHagimoto 0:0e0631af0305 133 @param nextPts output vector of 2D points (with single-precision floating-point coordinates)
RyoheiHagimoto 0:0e0631af0305 134 containing the calculated new positions of input features in the second image; when
RyoheiHagimoto 0:0e0631af0305 135 OPTFLOW_USE_INITIAL_FLOW flag is passed, the vector must have the same size as in the input.
RyoheiHagimoto 0:0e0631af0305 136 @param status output status vector (of unsigned chars); each element of the vector is set to 1 if
RyoheiHagimoto 0:0e0631af0305 137 the flow for the corresponding features has been found, otherwise, it is set to 0.
RyoheiHagimoto 0:0e0631af0305 138 @param err output vector of errors; each element of the vector is set to an error for the
RyoheiHagimoto 0:0e0631af0305 139 corresponding feature, type of the error measure can be set in flags parameter; if the flow wasn't
RyoheiHagimoto 0:0e0631af0305 140 found then the error is not defined (use the status parameter to find such cases).
RyoheiHagimoto 0:0e0631af0305 141 @param winSize size of the search window at each pyramid level.
RyoheiHagimoto 0:0e0631af0305 142 @param maxLevel 0-based maximal pyramid level number; if set to 0, pyramids are not used (single
RyoheiHagimoto 0:0e0631af0305 143 level), if set to 1, two levels are used, and so on; if pyramids are passed to input then
RyoheiHagimoto 0:0e0631af0305 144 algorithm will use as many levels as pyramids have but no more than maxLevel.
RyoheiHagimoto 0:0e0631af0305 145 @param criteria parameter, specifying the termination criteria of the iterative search algorithm
RyoheiHagimoto 0:0e0631af0305 146 (after the specified maximum number of iterations criteria.maxCount or when the search window
RyoheiHagimoto 0:0e0631af0305 147 moves by less than criteria.epsilon.
RyoheiHagimoto 0:0e0631af0305 148 @param flags operation flags:
RyoheiHagimoto 0:0e0631af0305 149 - **OPTFLOW_USE_INITIAL_FLOW** uses initial estimations, stored in nextPts; if the flag is
RyoheiHagimoto 0:0e0631af0305 150 not set, then prevPts is copied to nextPts and is considered the initial estimate.
RyoheiHagimoto 0:0e0631af0305 151 - **OPTFLOW_LK_GET_MIN_EIGENVALS** use minimum eigen values as an error measure (see
RyoheiHagimoto 0:0e0631af0305 152 minEigThreshold description); if the flag is not set, then L1 distance between patches
RyoheiHagimoto 0:0e0631af0305 153 around the original and a moved point, divided by number of pixels in a window, is used as a
RyoheiHagimoto 0:0e0631af0305 154 error measure.
RyoheiHagimoto 0:0e0631af0305 155 @param minEigThreshold the algorithm calculates the minimum eigen value of a 2x2 normal matrix of
RyoheiHagimoto 0:0e0631af0305 156 optical flow equations (this matrix is called a spatial gradient matrix in @cite Bouguet00), divided
RyoheiHagimoto 0:0e0631af0305 157 by number of pixels in a window; if this value is less than minEigThreshold, then a corresponding
RyoheiHagimoto 0:0e0631af0305 158 feature is filtered out and its flow is not processed, so it allows to remove bad points and get a
RyoheiHagimoto 0:0e0631af0305 159 performance boost.
RyoheiHagimoto 0:0e0631af0305 160
RyoheiHagimoto 0:0e0631af0305 161 The function implements a sparse iterative version of the Lucas-Kanade optical flow in pyramids. See
RyoheiHagimoto 0:0e0631af0305 162 @cite Bouguet00 . The function is parallelized with the TBB library.
RyoheiHagimoto 0:0e0631af0305 163
RyoheiHagimoto 0:0e0631af0305 164 @note
RyoheiHagimoto 0:0e0631af0305 165
RyoheiHagimoto 0:0e0631af0305 166 - An example using the Lucas-Kanade optical flow algorithm can be found at
RyoheiHagimoto 0:0e0631af0305 167 opencv_source_code/samples/cpp/lkdemo.cpp
RyoheiHagimoto 0:0e0631af0305 168 - (Python) An example using the Lucas-Kanade optical flow algorithm can be found at
RyoheiHagimoto 0:0e0631af0305 169 opencv_source_code/samples/python/lk_track.py
RyoheiHagimoto 0:0e0631af0305 170 - (Python) An example using the Lucas-Kanade tracker for homography matching can be found at
RyoheiHagimoto 0:0e0631af0305 171 opencv_source_code/samples/python/lk_homography.py
RyoheiHagimoto 0:0e0631af0305 172 */
RyoheiHagimoto 0:0e0631af0305 173 CV_EXPORTS_W void calcOpticalFlowPyrLK( InputArray prevImg, InputArray nextImg,
RyoheiHagimoto 0:0e0631af0305 174 InputArray prevPts, InputOutputArray nextPts,
RyoheiHagimoto 0:0e0631af0305 175 OutputArray status, OutputArray err,
RyoheiHagimoto 0:0e0631af0305 176 Size winSize = Size(21,21), int maxLevel = 3,
RyoheiHagimoto 0:0e0631af0305 177 TermCriteria criteria = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 30, 0.01),
RyoheiHagimoto 0:0e0631af0305 178 int flags = 0, double minEigThreshold = 1e-4 );
RyoheiHagimoto 0:0e0631af0305 179
RyoheiHagimoto 0:0e0631af0305 180 /** @brief Computes a dense optical flow using the Gunnar Farneback's algorithm.
RyoheiHagimoto 0:0e0631af0305 181
RyoheiHagimoto 0:0e0631af0305 182 @param prev first 8-bit single-channel input image.
RyoheiHagimoto 0:0e0631af0305 183 @param next second input image of the same size and the same type as prev.
RyoheiHagimoto 0:0e0631af0305 184 @param flow computed flow image that has the same size as prev and type CV_32FC2.
RyoheiHagimoto 0:0e0631af0305 185 @param pyr_scale parameter, specifying the image scale (\<1) to build pyramids for each image;
RyoheiHagimoto 0:0e0631af0305 186 pyr_scale=0.5 means a classical pyramid, where each next layer is twice smaller than the previous
RyoheiHagimoto 0:0e0631af0305 187 one.
RyoheiHagimoto 0:0e0631af0305 188 @param levels number of pyramid layers including the initial image; levels=1 means that no extra
RyoheiHagimoto 0:0e0631af0305 189 layers are created and only the original images are used.
RyoheiHagimoto 0:0e0631af0305 190 @param winsize averaging window size; larger values increase the algorithm robustness to image
RyoheiHagimoto 0:0e0631af0305 191 noise and give more chances for fast motion detection, but yield more blurred motion field.
RyoheiHagimoto 0:0e0631af0305 192 @param iterations number of iterations the algorithm does at each pyramid level.
RyoheiHagimoto 0:0e0631af0305 193 @param poly_n size of the pixel neighborhood used to find polynomial expansion in each pixel;
RyoheiHagimoto 0:0e0631af0305 194 larger values mean that the image will be approximated with smoother surfaces, yielding more
RyoheiHagimoto 0:0e0631af0305 195 robust algorithm and more blurred motion field, typically poly_n =5 or 7.
RyoheiHagimoto 0:0e0631af0305 196 @param poly_sigma standard deviation of the Gaussian that is used to smooth derivatives used as a
RyoheiHagimoto 0:0e0631af0305 197 basis for the polynomial expansion; for poly_n=5, you can set poly_sigma=1.1, for poly_n=7, a
RyoheiHagimoto 0:0e0631af0305 198 good value would be poly_sigma=1.5.
RyoheiHagimoto 0:0e0631af0305 199 @param flags operation flags that can be a combination of the following:
RyoheiHagimoto 0:0e0631af0305 200 - **OPTFLOW_USE_INITIAL_FLOW** uses the input flow as an initial flow approximation.
RyoheiHagimoto 0:0e0631af0305 201 - **OPTFLOW_FARNEBACK_GAUSSIAN** uses the Gaussian \f$\texttt{winsize}\times\texttt{winsize}\f$
RyoheiHagimoto 0:0e0631af0305 202 filter instead of a box filter of the same size for optical flow estimation; usually, this
RyoheiHagimoto 0:0e0631af0305 203 option gives z more accurate flow than with a box filter, at the cost of lower speed;
RyoheiHagimoto 0:0e0631af0305 204 normally, winsize for a Gaussian window should be set to a larger value to achieve the same
RyoheiHagimoto 0:0e0631af0305 205 level of robustness.
RyoheiHagimoto 0:0e0631af0305 206
RyoheiHagimoto 0:0e0631af0305 207 The function finds an optical flow for each prev pixel using the @cite Farneback2003 algorithm so that
RyoheiHagimoto 0:0e0631af0305 208
RyoheiHagimoto 0:0e0631af0305 209 \f[\texttt{prev} (y,x) \sim \texttt{next} ( y + \texttt{flow} (y,x)[1], x + \texttt{flow} (y,x)[0])\f]
RyoheiHagimoto 0:0e0631af0305 210
RyoheiHagimoto 0:0e0631af0305 211 @note
RyoheiHagimoto 0:0e0631af0305 212
RyoheiHagimoto 0:0e0631af0305 213 - An example using the optical flow algorithm described by Gunnar Farneback can be found at
RyoheiHagimoto 0:0e0631af0305 214 opencv_source_code/samples/cpp/fback.cpp
RyoheiHagimoto 0:0e0631af0305 215 - (Python) An example using the optical flow algorithm described by Gunnar Farneback can be
RyoheiHagimoto 0:0e0631af0305 216 found at opencv_source_code/samples/python/opt_flow.py
RyoheiHagimoto 0:0e0631af0305 217 */
RyoheiHagimoto 0:0e0631af0305 218 CV_EXPORTS_W void calcOpticalFlowFarneback( InputArray prev, InputArray next, InputOutputArray flow,
RyoheiHagimoto 0:0e0631af0305 219 double pyr_scale, int levels, int winsize,
RyoheiHagimoto 0:0e0631af0305 220 int iterations, int poly_n, double poly_sigma,
RyoheiHagimoto 0:0e0631af0305 221 int flags );
RyoheiHagimoto 0:0e0631af0305 222
RyoheiHagimoto 0:0e0631af0305 223 /** @brief Computes an optimal affine transformation between two 2D point sets.
RyoheiHagimoto 0:0e0631af0305 224
RyoheiHagimoto 0:0e0631af0305 225 @param src First input 2D point set stored in std::vector or Mat, or an image stored in Mat.
RyoheiHagimoto 0:0e0631af0305 226 @param dst Second input 2D point set of the same size and the same type as A, or another image.
RyoheiHagimoto 0:0e0631af0305 227 @param fullAffine If true, the function finds an optimal affine transformation with no additional
RyoheiHagimoto 0:0e0631af0305 228 restrictions (6 degrees of freedom). Otherwise, the class of transformations to choose from is
RyoheiHagimoto 0:0e0631af0305 229 limited to combinations of translation, rotation, and uniform scaling (4 degrees of freedom).
RyoheiHagimoto 0:0e0631af0305 230
RyoheiHagimoto 0:0e0631af0305 231 The function finds an optimal affine transform *[A|b]* (a 2 x 3 floating-point matrix) that
RyoheiHagimoto 0:0e0631af0305 232 approximates best the affine transformation between:
RyoheiHagimoto 0:0e0631af0305 233
RyoheiHagimoto 0:0e0631af0305 234 * Two point sets
RyoheiHagimoto 0:0e0631af0305 235 * Two raster images. In this case, the function first finds some features in the src image and
RyoheiHagimoto 0:0e0631af0305 236 finds the corresponding features in dst image. After that, the problem is reduced to the first
RyoheiHagimoto 0:0e0631af0305 237 case.
RyoheiHagimoto 0:0e0631af0305 238 In case of point sets, the problem is formulated as follows: you need to find a 2x2 matrix *A* and
RyoheiHagimoto 0:0e0631af0305 239 2x1 vector *b* so that:
RyoheiHagimoto 0:0e0631af0305 240
RyoheiHagimoto 0:0e0631af0305 241 \f[[A^*|b^*] = arg \min _{[A|b]} \sum _i \| \texttt{dst}[i] - A { \texttt{src}[i]}^T - b \| ^2\f]
RyoheiHagimoto 0:0e0631af0305 242 where src[i] and dst[i] are the i-th points in src and dst, respectively
RyoheiHagimoto 0:0e0631af0305 243 \f$[A|b]\f$ can be either arbitrary (when fullAffine=true ) or have a form of
RyoheiHagimoto 0:0e0631af0305 244 \f[\begin{bmatrix} a_{11} & a_{12} & b_1 \\ -a_{12} & a_{11} & b_2 \end{bmatrix}\f]
RyoheiHagimoto 0:0e0631af0305 245 when fullAffine=false.
RyoheiHagimoto 0:0e0631af0305 246
RyoheiHagimoto 0:0e0631af0305 247 @sa
RyoheiHagimoto 0:0e0631af0305 248 estimateAffine2D, estimateAffinePartial2D, getAffineTransform, getPerspectiveTransform, findHomography
RyoheiHagimoto 0:0e0631af0305 249 */
RyoheiHagimoto 0:0e0631af0305 250 CV_EXPORTS_W Mat estimateRigidTransform( InputArray src, InputArray dst, bool fullAffine );
RyoheiHagimoto 0:0e0631af0305 251
RyoheiHagimoto 0:0e0631af0305 252
RyoheiHagimoto 0:0e0631af0305 253 enum
RyoheiHagimoto 0:0e0631af0305 254 {
RyoheiHagimoto 0:0e0631af0305 255 MOTION_TRANSLATION = 0,
RyoheiHagimoto 0:0e0631af0305 256 MOTION_EUCLIDEAN = 1,
RyoheiHagimoto 0:0e0631af0305 257 MOTION_AFFINE = 2,
RyoheiHagimoto 0:0e0631af0305 258 MOTION_HOMOGRAPHY = 3
RyoheiHagimoto 0:0e0631af0305 259 };
RyoheiHagimoto 0:0e0631af0305 260
RyoheiHagimoto 0:0e0631af0305 261 /** @brief Finds the geometric transform (warp) between two images in terms of the ECC criterion @cite EP08 .
RyoheiHagimoto 0:0e0631af0305 262
RyoheiHagimoto 0:0e0631af0305 263 @param templateImage single-channel template image; CV_8U or CV_32F array.
RyoheiHagimoto 0:0e0631af0305 264 @param inputImage single-channel input image which should be warped with the final warpMatrix in
RyoheiHagimoto 0:0e0631af0305 265 order to provide an image similar to templateImage, same type as temlateImage.
RyoheiHagimoto 0:0e0631af0305 266 @param warpMatrix floating-point \f$2\times 3\f$ or \f$3\times 3\f$ mapping matrix (warp).
RyoheiHagimoto 0:0e0631af0305 267 @param motionType parameter, specifying the type of motion:
RyoheiHagimoto 0:0e0631af0305 268 - **MOTION_TRANSLATION** sets a translational motion model; warpMatrix is \f$2\times 3\f$ with
RyoheiHagimoto 0:0e0631af0305 269 the first \f$2\times 2\f$ part being the unity matrix and the rest two parameters being
RyoheiHagimoto 0:0e0631af0305 270 estimated.
RyoheiHagimoto 0:0e0631af0305 271 - **MOTION_EUCLIDEAN** sets a Euclidean (rigid) transformation as motion model; three
RyoheiHagimoto 0:0e0631af0305 272 parameters are estimated; warpMatrix is \f$2\times 3\f$.
RyoheiHagimoto 0:0e0631af0305 273 - **MOTION_AFFINE** sets an affine motion model (DEFAULT); six parameters are estimated;
RyoheiHagimoto 0:0e0631af0305 274 warpMatrix is \f$2\times 3\f$.
RyoheiHagimoto 0:0e0631af0305 275 - **MOTION_HOMOGRAPHY** sets a homography as a motion model; eight parameters are
RyoheiHagimoto 0:0e0631af0305 276 estimated;\`warpMatrix\` is \f$3\times 3\f$.
RyoheiHagimoto 0:0e0631af0305 277 @param criteria parameter, specifying the termination criteria of the ECC algorithm;
RyoheiHagimoto 0:0e0631af0305 278 criteria.epsilon defines the threshold of the increment in the correlation coefficient between two
RyoheiHagimoto 0:0e0631af0305 279 iterations (a negative criteria.epsilon makes criteria.maxcount the only termination criterion).
RyoheiHagimoto 0:0e0631af0305 280 Default values are shown in the declaration above.
RyoheiHagimoto 0:0e0631af0305 281 @param inputMask An optional mask to indicate valid values of inputImage.
RyoheiHagimoto 0:0e0631af0305 282
RyoheiHagimoto 0:0e0631af0305 283 The function estimates the optimum transformation (warpMatrix) with respect to ECC criterion
RyoheiHagimoto 0:0e0631af0305 284 (@cite EP08), that is
RyoheiHagimoto 0:0e0631af0305 285
RyoheiHagimoto 0:0e0631af0305 286 \f[\texttt{warpMatrix} = \texttt{warpMatrix} = \arg\max_{W} \texttt{ECC}(\texttt{templateImage}(x,y),\texttt{inputImage}(x',y'))\f]
RyoheiHagimoto 0:0e0631af0305 287
RyoheiHagimoto 0:0e0631af0305 288 where
RyoheiHagimoto 0:0e0631af0305 289
RyoheiHagimoto 0:0e0631af0305 290 \f[\begin{bmatrix} x' \\ y' \end{bmatrix} = W \cdot \begin{bmatrix} x \\ y \\ 1 \end{bmatrix}\f]
RyoheiHagimoto 0:0e0631af0305 291
RyoheiHagimoto 0:0e0631af0305 292 (the equation holds with homogeneous coordinates for homography). It returns the final enhanced
RyoheiHagimoto 0:0e0631af0305 293 correlation coefficient, that is the correlation coefficient between the template image and the
RyoheiHagimoto 0:0e0631af0305 294 final warped input image. When a \f$3\times 3\f$ matrix is given with motionType =0, 1 or 2, the third
RyoheiHagimoto 0:0e0631af0305 295 row is ignored.
RyoheiHagimoto 0:0e0631af0305 296
RyoheiHagimoto 0:0e0631af0305 297 Unlike findHomography and estimateRigidTransform, the function findTransformECC implements an
RyoheiHagimoto 0:0e0631af0305 298 area-based alignment that builds on intensity similarities. In essence, the function updates the
RyoheiHagimoto 0:0e0631af0305 299 initial transformation that roughly aligns the images. If this information is missing, the identity
RyoheiHagimoto 0:0e0631af0305 300 warp (unity matrix) should be given as input. Note that if images undergo strong
RyoheiHagimoto 0:0e0631af0305 301 displacements/rotations, an initial transformation that roughly aligns the images is necessary
RyoheiHagimoto 0:0e0631af0305 302 (e.g., a simple euclidean/similarity transform that allows for the images showing the same image
RyoheiHagimoto 0:0e0631af0305 303 content approximately). Use inverse warping in the second image to take an image close to the first
RyoheiHagimoto 0:0e0631af0305 304 one, i.e. use the flag WARP_INVERSE_MAP with warpAffine or warpPerspective. See also the OpenCV
RyoheiHagimoto 0:0e0631af0305 305 sample image_alignment.cpp that demonstrates the use of the function. Note that the function throws
RyoheiHagimoto 0:0e0631af0305 306 an exception if algorithm does not converges.
RyoheiHagimoto 0:0e0631af0305 307
RyoheiHagimoto 0:0e0631af0305 308 @sa
RyoheiHagimoto 0:0e0631af0305 309 estimateAffine2D, estimateAffinePartial2D, findHomography
RyoheiHagimoto 0:0e0631af0305 310 */
RyoheiHagimoto 0:0e0631af0305 311 CV_EXPORTS_W double findTransformECC( InputArray templateImage, InputArray inputImage,
RyoheiHagimoto 0:0e0631af0305 312 InputOutputArray warpMatrix, int motionType = MOTION_AFFINE,
RyoheiHagimoto 0:0e0631af0305 313 TermCriteria criteria = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 50, 0.001),
RyoheiHagimoto 0:0e0631af0305 314 InputArray inputMask = noArray());
RyoheiHagimoto 0:0e0631af0305 315
RyoheiHagimoto 0:0e0631af0305 316 /** @brief Kalman filter class.
RyoheiHagimoto 0:0e0631af0305 317
RyoheiHagimoto 0:0e0631af0305 318 The class implements a standard Kalman filter <http://en.wikipedia.org/wiki/Kalman_filter>,
RyoheiHagimoto 0:0e0631af0305 319 @cite Welch95 . However, you can modify transitionMatrix, controlMatrix, and measurementMatrix to get
RyoheiHagimoto 0:0e0631af0305 320 an extended Kalman filter functionality. See the OpenCV sample kalman.cpp.
RyoheiHagimoto 0:0e0631af0305 321
RyoheiHagimoto 0:0e0631af0305 322 @note
RyoheiHagimoto 0:0e0631af0305 323
RyoheiHagimoto 0:0e0631af0305 324 - An example using the standard Kalman filter can be found at
RyoheiHagimoto 0:0e0631af0305 325 opencv_source_code/samples/cpp/kalman.cpp
RyoheiHagimoto 0:0e0631af0305 326 */
RyoheiHagimoto 0:0e0631af0305 327 class CV_EXPORTS_W KalmanFilter
RyoheiHagimoto 0:0e0631af0305 328 {
RyoheiHagimoto 0:0e0631af0305 329 public:
RyoheiHagimoto 0:0e0631af0305 330 /** @brief The constructors.
RyoheiHagimoto 0:0e0631af0305 331
RyoheiHagimoto 0:0e0631af0305 332 @note In C API when CvKalman\* kalmanFilter structure is not needed anymore, it should be released
RyoheiHagimoto 0:0e0631af0305 333 with cvReleaseKalman(&kalmanFilter)
RyoheiHagimoto 0:0e0631af0305 334 */
RyoheiHagimoto 0:0e0631af0305 335 CV_WRAP KalmanFilter();
RyoheiHagimoto 0:0e0631af0305 336 /** @overload
RyoheiHagimoto 0:0e0631af0305 337 @param dynamParams Dimensionality of the state.
RyoheiHagimoto 0:0e0631af0305 338 @param measureParams Dimensionality of the measurement.
RyoheiHagimoto 0:0e0631af0305 339 @param controlParams Dimensionality of the control vector.
RyoheiHagimoto 0:0e0631af0305 340 @param type Type of the created matrices that should be CV_32F or CV_64F.
RyoheiHagimoto 0:0e0631af0305 341 */
RyoheiHagimoto 0:0e0631af0305 342 CV_WRAP KalmanFilter( int dynamParams, int measureParams, int controlParams = 0, int type = CV_32F );
RyoheiHagimoto 0:0e0631af0305 343
RyoheiHagimoto 0:0e0631af0305 344 /** @brief Re-initializes Kalman filter. The previous content is destroyed.
RyoheiHagimoto 0:0e0631af0305 345
RyoheiHagimoto 0:0e0631af0305 346 @param dynamParams Dimensionality of the state.
RyoheiHagimoto 0:0e0631af0305 347 @param measureParams Dimensionality of the measurement.
RyoheiHagimoto 0:0e0631af0305 348 @param controlParams Dimensionality of the control vector.
RyoheiHagimoto 0:0e0631af0305 349 @param type Type of the created matrices that should be CV_32F or CV_64F.
RyoheiHagimoto 0:0e0631af0305 350 */
RyoheiHagimoto 0:0e0631af0305 351 void init( int dynamParams, int measureParams, int controlParams = 0, int type = CV_32F );
RyoheiHagimoto 0:0e0631af0305 352
RyoheiHagimoto 0:0e0631af0305 353 /** @brief Computes a predicted state.
RyoheiHagimoto 0:0e0631af0305 354
RyoheiHagimoto 0:0e0631af0305 355 @param control The optional input control
RyoheiHagimoto 0:0e0631af0305 356 */
RyoheiHagimoto 0:0e0631af0305 357 CV_WRAP const Mat& predict( const Mat& control = Mat() );
RyoheiHagimoto 0:0e0631af0305 358
RyoheiHagimoto 0:0e0631af0305 359 /** @brief Updates the predicted state from the measurement.
RyoheiHagimoto 0:0e0631af0305 360
RyoheiHagimoto 0:0e0631af0305 361 @param measurement The measured system parameters
RyoheiHagimoto 0:0e0631af0305 362 */
RyoheiHagimoto 0:0e0631af0305 363 CV_WRAP const Mat& correct( const Mat& measurement );
RyoheiHagimoto 0:0e0631af0305 364
RyoheiHagimoto 0:0e0631af0305 365 CV_PROP_RW Mat statePre; //!< predicted state (x'(k)): x(k)=A*x(k-1)+B*u(k)
RyoheiHagimoto 0:0e0631af0305 366 CV_PROP_RW Mat statePost; //!< corrected state (x(k)): x(k)=x'(k)+K(k)*(z(k)-H*x'(k))
RyoheiHagimoto 0:0e0631af0305 367 CV_PROP_RW Mat transitionMatrix; //!< state transition matrix (A)
RyoheiHagimoto 0:0e0631af0305 368 CV_PROP_RW Mat controlMatrix; //!< control matrix (B) (not used if there is no control)
RyoheiHagimoto 0:0e0631af0305 369 CV_PROP_RW Mat measurementMatrix; //!< measurement matrix (H)
RyoheiHagimoto 0:0e0631af0305 370 CV_PROP_RW Mat processNoiseCov; //!< process noise covariance matrix (Q)
RyoheiHagimoto 0:0e0631af0305 371 CV_PROP_RW Mat measurementNoiseCov;//!< measurement noise covariance matrix (R)
RyoheiHagimoto 0:0e0631af0305 372 CV_PROP_RW Mat errorCovPre; //!< priori error estimate covariance matrix (P'(k)): P'(k)=A*P(k-1)*At + Q)*/
RyoheiHagimoto 0:0e0631af0305 373 CV_PROP_RW Mat gain; //!< Kalman gain matrix (K(k)): K(k)=P'(k)*Ht*inv(H*P'(k)*Ht+R)
RyoheiHagimoto 0:0e0631af0305 374 CV_PROP_RW Mat errorCovPost; //!< posteriori error estimate covariance matrix (P(k)): P(k)=(I-K(k)*H)*P'(k)
RyoheiHagimoto 0:0e0631af0305 375
RyoheiHagimoto 0:0e0631af0305 376 // temporary matrices
RyoheiHagimoto 0:0e0631af0305 377 Mat temp1;
RyoheiHagimoto 0:0e0631af0305 378 Mat temp2;
RyoheiHagimoto 0:0e0631af0305 379 Mat temp3;
RyoheiHagimoto 0:0e0631af0305 380 Mat temp4;
RyoheiHagimoto 0:0e0631af0305 381 Mat temp5;
RyoheiHagimoto 0:0e0631af0305 382 };
RyoheiHagimoto 0:0e0631af0305 383
RyoheiHagimoto 0:0e0631af0305 384
RyoheiHagimoto 0:0e0631af0305 385 class CV_EXPORTS_W DenseOpticalFlow : public Algorithm
RyoheiHagimoto 0:0e0631af0305 386 {
RyoheiHagimoto 0:0e0631af0305 387 public:
RyoheiHagimoto 0:0e0631af0305 388 /** @brief Calculates an optical flow.
RyoheiHagimoto 0:0e0631af0305 389
RyoheiHagimoto 0:0e0631af0305 390 @param I0 first 8-bit single-channel input image.
RyoheiHagimoto 0:0e0631af0305 391 @param I1 second input image of the same size and the same type as prev.
RyoheiHagimoto 0:0e0631af0305 392 @param flow computed flow image that has the same size as prev and type CV_32FC2.
RyoheiHagimoto 0:0e0631af0305 393 */
RyoheiHagimoto 0:0e0631af0305 394 CV_WRAP virtual void calc( InputArray I0, InputArray I1, InputOutputArray flow ) = 0;
RyoheiHagimoto 0:0e0631af0305 395 /** @brief Releases all inner buffers.
RyoheiHagimoto 0:0e0631af0305 396 */
RyoheiHagimoto 0:0e0631af0305 397 CV_WRAP virtual void collectGarbage() = 0;
RyoheiHagimoto 0:0e0631af0305 398 };
RyoheiHagimoto 0:0e0631af0305 399
RyoheiHagimoto 0:0e0631af0305 400 /** @brief Base interface for sparse optical flow algorithms.
RyoheiHagimoto 0:0e0631af0305 401 */
RyoheiHagimoto 0:0e0631af0305 402 class CV_EXPORTS_W SparseOpticalFlow : public Algorithm
RyoheiHagimoto 0:0e0631af0305 403 {
RyoheiHagimoto 0:0e0631af0305 404 public:
RyoheiHagimoto 0:0e0631af0305 405 /** @brief Calculates a sparse optical flow.
RyoheiHagimoto 0:0e0631af0305 406
RyoheiHagimoto 0:0e0631af0305 407 @param prevImg First input image.
RyoheiHagimoto 0:0e0631af0305 408 @param nextImg Second input image of the same size and the same type as prevImg.
RyoheiHagimoto 0:0e0631af0305 409 @param prevPts Vector of 2D points for which the flow needs to be found.
RyoheiHagimoto 0:0e0631af0305 410 @param nextPts Output vector of 2D points containing the calculated new positions of input features in the second image.
RyoheiHagimoto 0:0e0631af0305 411 @param status Output status vector. Each element of the vector is set to 1 if the
RyoheiHagimoto 0:0e0631af0305 412 flow for the corresponding features has been found. Otherwise, it is set to 0.
RyoheiHagimoto 0:0e0631af0305 413 @param err Optional output vector that contains error response for each point (inverse confidence).
RyoheiHagimoto 0:0e0631af0305 414 */
RyoheiHagimoto 0:0e0631af0305 415 CV_WRAP virtual void calc(InputArray prevImg, InputArray nextImg,
RyoheiHagimoto 0:0e0631af0305 416 InputArray prevPts, InputOutputArray nextPts,
RyoheiHagimoto 0:0e0631af0305 417 OutputArray status,
RyoheiHagimoto 0:0e0631af0305 418 OutputArray err = cv::noArray()) = 0;
RyoheiHagimoto 0:0e0631af0305 419 };
RyoheiHagimoto 0:0e0631af0305 420
RyoheiHagimoto 0:0e0631af0305 421 /** @brief "Dual TV L1" Optical Flow Algorithm.
RyoheiHagimoto 0:0e0631af0305 422
RyoheiHagimoto 0:0e0631af0305 423 The class implements the "Dual TV L1" optical flow algorithm described in @cite Zach2007 and
RyoheiHagimoto 0:0e0631af0305 424 @cite Javier2012 .
RyoheiHagimoto 0:0e0631af0305 425 Here are important members of the class that control the algorithm, which you can set after
RyoheiHagimoto 0:0e0631af0305 426 constructing the class instance:
RyoheiHagimoto 0:0e0631af0305 427
RyoheiHagimoto 0:0e0631af0305 428 - member double tau
RyoheiHagimoto 0:0e0631af0305 429 Time step of the numerical scheme.
RyoheiHagimoto 0:0e0631af0305 430
RyoheiHagimoto 0:0e0631af0305 431 - member double lambda
RyoheiHagimoto 0:0e0631af0305 432 Weight parameter for the data term, attachment parameter. This is the most relevant
RyoheiHagimoto 0:0e0631af0305 433 parameter, which determines the smoothness of the output. The smaller this parameter is,
RyoheiHagimoto 0:0e0631af0305 434 the smoother the solutions we obtain. It depends on the range of motions of the images, so
RyoheiHagimoto 0:0e0631af0305 435 its value should be adapted to each image sequence.
RyoheiHagimoto 0:0e0631af0305 436
RyoheiHagimoto 0:0e0631af0305 437 - member double theta
RyoheiHagimoto 0:0e0631af0305 438 Weight parameter for (u - v)\^2, tightness parameter. It serves as a link between the
RyoheiHagimoto 0:0e0631af0305 439 attachment and the regularization terms. In theory, it should have a small value in order
RyoheiHagimoto 0:0e0631af0305 440 to maintain both parts in correspondence. The method is stable for a large range of values
RyoheiHagimoto 0:0e0631af0305 441 of this parameter.
RyoheiHagimoto 0:0e0631af0305 442
RyoheiHagimoto 0:0e0631af0305 443 - member int nscales
RyoheiHagimoto 0:0e0631af0305 444 Number of scales used to create the pyramid of images.
RyoheiHagimoto 0:0e0631af0305 445
RyoheiHagimoto 0:0e0631af0305 446 - member int warps
RyoheiHagimoto 0:0e0631af0305 447 Number of warpings per scale. Represents the number of times that I1(x+u0) and grad(
RyoheiHagimoto 0:0e0631af0305 448 I1(x+u0) ) are computed per scale. This is a parameter that assures the stability of the
RyoheiHagimoto 0:0e0631af0305 449 method. It also affects the running time, so it is a compromise between speed and
RyoheiHagimoto 0:0e0631af0305 450 accuracy.
RyoheiHagimoto 0:0e0631af0305 451
RyoheiHagimoto 0:0e0631af0305 452 - member double epsilon
RyoheiHagimoto 0:0e0631af0305 453 Stopping criterion threshold used in the numerical scheme, which is a trade-off between
RyoheiHagimoto 0:0e0631af0305 454 precision and running time. A small value will yield more accurate solutions at the
RyoheiHagimoto 0:0e0631af0305 455 expense of a slower convergence.
RyoheiHagimoto 0:0e0631af0305 456
RyoheiHagimoto 0:0e0631af0305 457 - member int iterations
RyoheiHagimoto 0:0e0631af0305 458 Stopping criterion iterations number used in the numerical scheme.
RyoheiHagimoto 0:0e0631af0305 459
RyoheiHagimoto 0:0e0631af0305 460 C. Zach, T. Pock and H. Bischof, "A Duality Based Approach for Realtime TV-L1 Optical Flow".
RyoheiHagimoto 0:0e0631af0305 461 Javier Sanchez, Enric Meinhardt-Llopis and Gabriele Facciolo. "TV-L1 Optical Flow Estimation".
RyoheiHagimoto 0:0e0631af0305 462 */
RyoheiHagimoto 0:0e0631af0305 463 class CV_EXPORTS_W DualTVL1OpticalFlow : public DenseOpticalFlow
RyoheiHagimoto 0:0e0631af0305 464 {
RyoheiHagimoto 0:0e0631af0305 465 public:
RyoheiHagimoto 0:0e0631af0305 466 //! @brief Time step of the numerical scheme
RyoheiHagimoto 0:0e0631af0305 467 /** @see setTau */
RyoheiHagimoto 0:0e0631af0305 468 CV_WRAP virtual double getTau() const = 0;
RyoheiHagimoto 0:0e0631af0305 469 /** @copybrief getTau @see getTau */
RyoheiHagimoto 0:0e0631af0305 470 CV_WRAP virtual void setTau(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 471 //! @brief Weight parameter for the data term, attachment parameter
RyoheiHagimoto 0:0e0631af0305 472 /** @see setLambda */
RyoheiHagimoto 0:0e0631af0305 473 CV_WRAP virtual double getLambda() const = 0;
RyoheiHagimoto 0:0e0631af0305 474 /** @copybrief getLambda @see getLambda */
RyoheiHagimoto 0:0e0631af0305 475 CV_WRAP virtual void setLambda(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 476 //! @brief Weight parameter for (u - v)^2, tightness parameter
RyoheiHagimoto 0:0e0631af0305 477 /** @see setTheta */
RyoheiHagimoto 0:0e0631af0305 478 CV_WRAP virtual double getTheta() const = 0;
RyoheiHagimoto 0:0e0631af0305 479 /** @copybrief getTheta @see getTheta */
RyoheiHagimoto 0:0e0631af0305 480 CV_WRAP virtual void setTheta(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 481 //! @brief coefficient for additional illumination variation term
RyoheiHagimoto 0:0e0631af0305 482 /** @see setGamma */
RyoheiHagimoto 0:0e0631af0305 483 CV_WRAP virtual double getGamma() const = 0;
RyoheiHagimoto 0:0e0631af0305 484 /** @copybrief getGamma @see getGamma */
RyoheiHagimoto 0:0e0631af0305 485 CV_WRAP virtual void setGamma(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 486 //! @brief Number of scales used to create the pyramid of images
RyoheiHagimoto 0:0e0631af0305 487 /** @see setScalesNumber */
RyoheiHagimoto 0:0e0631af0305 488 CV_WRAP virtual int getScalesNumber() const = 0;
RyoheiHagimoto 0:0e0631af0305 489 /** @copybrief getScalesNumber @see getScalesNumber */
RyoheiHagimoto 0:0e0631af0305 490 CV_WRAP virtual void setScalesNumber(int val) = 0;
RyoheiHagimoto 0:0e0631af0305 491 //! @brief Number of warpings per scale
RyoheiHagimoto 0:0e0631af0305 492 /** @see setWarpingsNumber */
RyoheiHagimoto 0:0e0631af0305 493 CV_WRAP virtual int getWarpingsNumber() const = 0;
RyoheiHagimoto 0:0e0631af0305 494 /** @copybrief getWarpingsNumber @see getWarpingsNumber */
RyoheiHagimoto 0:0e0631af0305 495 CV_WRAP virtual void setWarpingsNumber(int val) = 0;
RyoheiHagimoto 0:0e0631af0305 496 //! @brief Stopping criterion threshold used in the numerical scheme, which is a trade-off between precision and running time
RyoheiHagimoto 0:0e0631af0305 497 /** @see setEpsilon */
RyoheiHagimoto 0:0e0631af0305 498 CV_WRAP virtual double getEpsilon() const = 0;
RyoheiHagimoto 0:0e0631af0305 499 /** @copybrief getEpsilon @see getEpsilon */
RyoheiHagimoto 0:0e0631af0305 500 CV_WRAP virtual void setEpsilon(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 501 //! @brief Inner iterations (between outlier filtering) used in the numerical scheme
RyoheiHagimoto 0:0e0631af0305 502 /** @see setInnerIterations */
RyoheiHagimoto 0:0e0631af0305 503 CV_WRAP virtual int getInnerIterations() const = 0;
RyoheiHagimoto 0:0e0631af0305 504 /** @copybrief getInnerIterations @see getInnerIterations */
RyoheiHagimoto 0:0e0631af0305 505 CV_WRAP virtual void setInnerIterations(int val) = 0;
RyoheiHagimoto 0:0e0631af0305 506 //! @brief Outer iterations (number of inner loops) used in the numerical scheme
RyoheiHagimoto 0:0e0631af0305 507 /** @see setOuterIterations */
RyoheiHagimoto 0:0e0631af0305 508 CV_WRAP virtual int getOuterIterations() const = 0;
RyoheiHagimoto 0:0e0631af0305 509 /** @copybrief getOuterIterations @see getOuterIterations */
RyoheiHagimoto 0:0e0631af0305 510 CV_WRAP virtual void setOuterIterations(int val) = 0;
RyoheiHagimoto 0:0e0631af0305 511 //! @brief Use initial flow
RyoheiHagimoto 0:0e0631af0305 512 /** @see setUseInitialFlow */
RyoheiHagimoto 0:0e0631af0305 513 CV_WRAP virtual bool getUseInitialFlow() const = 0;
RyoheiHagimoto 0:0e0631af0305 514 /** @copybrief getUseInitialFlow @see getUseInitialFlow */
RyoheiHagimoto 0:0e0631af0305 515 CV_WRAP virtual void setUseInitialFlow(bool val) = 0;
RyoheiHagimoto 0:0e0631af0305 516 //! @brief Step between scales (<1)
RyoheiHagimoto 0:0e0631af0305 517 /** @see setScaleStep */
RyoheiHagimoto 0:0e0631af0305 518 CV_WRAP virtual double getScaleStep() const = 0;
RyoheiHagimoto 0:0e0631af0305 519 /** @copybrief getScaleStep @see getScaleStep */
RyoheiHagimoto 0:0e0631af0305 520 CV_WRAP virtual void setScaleStep(double val) = 0;
RyoheiHagimoto 0:0e0631af0305 521 //! @brief Median filter kernel size (1 = no filter) (3 or 5)
RyoheiHagimoto 0:0e0631af0305 522 /** @see setMedianFiltering */
RyoheiHagimoto 0:0e0631af0305 523 CV_WRAP virtual int getMedianFiltering() const = 0;
RyoheiHagimoto 0:0e0631af0305 524 /** @copybrief getMedianFiltering @see getMedianFiltering */
RyoheiHagimoto 0:0e0631af0305 525 CV_WRAP virtual void setMedianFiltering(int val) = 0;
RyoheiHagimoto 0:0e0631af0305 526
RyoheiHagimoto 0:0e0631af0305 527 /** @brief Creates instance of cv::DualTVL1OpticalFlow*/
RyoheiHagimoto 0:0e0631af0305 528 CV_WRAP static Ptr<DualTVL1OpticalFlow> create(
RyoheiHagimoto 0:0e0631af0305 529 double tau = 0.25,
RyoheiHagimoto 0:0e0631af0305 530 double lambda = 0.15,
RyoheiHagimoto 0:0e0631af0305 531 double theta = 0.3,
RyoheiHagimoto 0:0e0631af0305 532 int nscales = 5,
RyoheiHagimoto 0:0e0631af0305 533 int warps = 5,
RyoheiHagimoto 0:0e0631af0305 534 double epsilon = 0.01,
RyoheiHagimoto 0:0e0631af0305 535 int innnerIterations = 30,
RyoheiHagimoto 0:0e0631af0305 536 int outerIterations = 10,
RyoheiHagimoto 0:0e0631af0305 537 double scaleStep = 0.8,
RyoheiHagimoto 0:0e0631af0305 538 double gamma = 0.0,
RyoheiHagimoto 0:0e0631af0305 539 int medianFiltering = 5,
RyoheiHagimoto 0:0e0631af0305 540 bool useInitialFlow = false);
RyoheiHagimoto 0:0e0631af0305 541 };
RyoheiHagimoto 0:0e0631af0305 542
RyoheiHagimoto 0:0e0631af0305 543 /** @brief Creates instance of cv::DenseOpticalFlow
RyoheiHagimoto 0:0e0631af0305 544 */
RyoheiHagimoto 0:0e0631af0305 545 CV_EXPORTS_W Ptr<DualTVL1OpticalFlow> createOptFlow_DualTVL1();
RyoheiHagimoto 0:0e0631af0305 546
RyoheiHagimoto 0:0e0631af0305 547 /** @brief Class computing a dense optical flow using the Gunnar Farneback’s algorithm.
RyoheiHagimoto 0:0e0631af0305 548 */
RyoheiHagimoto 0:0e0631af0305 549 class CV_EXPORTS_W FarnebackOpticalFlow : public DenseOpticalFlow
RyoheiHagimoto 0:0e0631af0305 550 {
RyoheiHagimoto 0:0e0631af0305 551 public:
RyoheiHagimoto 0:0e0631af0305 552 CV_WRAP virtual int getNumLevels() const = 0;
RyoheiHagimoto 0:0e0631af0305 553 CV_WRAP virtual void setNumLevels(int numLevels) = 0;
RyoheiHagimoto 0:0e0631af0305 554
RyoheiHagimoto 0:0e0631af0305 555 CV_WRAP virtual double getPyrScale() const = 0;
RyoheiHagimoto 0:0e0631af0305 556 CV_WRAP virtual void setPyrScale(double pyrScale) = 0;
RyoheiHagimoto 0:0e0631af0305 557
RyoheiHagimoto 0:0e0631af0305 558 CV_WRAP virtual bool getFastPyramids() const = 0;
RyoheiHagimoto 0:0e0631af0305 559 CV_WRAP virtual void setFastPyramids(bool fastPyramids) = 0;
RyoheiHagimoto 0:0e0631af0305 560
RyoheiHagimoto 0:0e0631af0305 561 CV_WRAP virtual int getWinSize() const = 0;
RyoheiHagimoto 0:0e0631af0305 562 CV_WRAP virtual void setWinSize(int winSize) = 0;
RyoheiHagimoto 0:0e0631af0305 563
RyoheiHagimoto 0:0e0631af0305 564 CV_WRAP virtual int getNumIters() const = 0;
RyoheiHagimoto 0:0e0631af0305 565 CV_WRAP virtual void setNumIters(int numIters) = 0;
RyoheiHagimoto 0:0e0631af0305 566
RyoheiHagimoto 0:0e0631af0305 567 CV_WRAP virtual int getPolyN() const = 0;
RyoheiHagimoto 0:0e0631af0305 568 CV_WRAP virtual void setPolyN(int polyN) = 0;
RyoheiHagimoto 0:0e0631af0305 569
RyoheiHagimoto 0:0e0631af0305 570 CV_WRAP virtual double getPolySigma() const = 0;
RyoheiHagimoto 0:0e0631af0305 571 CV_WRAP virtual void setPolySigma(double polySigma) = 0;
RyoheiHagimoto 0:0e0631af0305 572
RyoheiHagimoto 0:0e0631af0305 573 CV_WRAP virtual int getFlags() const = 0;
RyoheiHagimoto 0:0e0631af0305 574 CV_WRAP virtual void setFlags(int flags) = 0;
RyoheiHagimoto 0:0e0631af0305 575
RyoheiHagimoto 0:0e0631af0305 576 CV_WRAP static Ptr<FarnebackOpticalFlow> create(
RyoheiHagimoto 0:0e0631af0305 577 int numLevels = 5,
RyoheiHagimoto 0:0e0631af0305 578 double pyrScale = 0.5,
RyoheiHagimoto 0:0e0631af0305 579 bool fastPyramids = false,
RyoheiHagimoto 0:0e0631af0305 580 int winSize = 13,
RyoheiHagimoto 0:0e0631af0305 581 int numIters = 10,
RyoheiHagimoto 0:0e0631af0305 582 int polyN = 5,
RyoheiHagimoto 0:0e0631af0305 583 double polySigma = 1.1,
RyoheiHagimoto 0:0e0631af0305 584 int flags = 0);
RyoheiHagimoto 0:0e0631af0305 585 };
RyoheiHagimoto 0:0e0631af0305 586
RyoheiHagimoto 0:0e0631af0305 587
RyoheiHagimoto 0:0e0631af0305 588 /** @brief Class used for calculating a sparse optical flow.
RyoheiHagimoto 0:0e0631af0305 589
RyoheiHagimoto 0:0e0631af0305 590 The class can calculate an optical flow for a sparse feature set using the
RyoheiHagimoto 0:0e0631af0305 591 iterative Lucas-Kanade method with pyramids.
RyoheiHagimoto 0:0e0631af0305 592
RyoheiHagimoto 0:0e0631af0305 593 @sa calcOpticalFlowPyrLK
RyoheiHagimoto 0:0e0631af0305 594
RyoheiHagimoto 0:0e0631af0305 595 */
RyoheiHagimoto 0:0e0631af0305 596 class CV_EXPORTS_W SparsePyrLKOpticalFlow : public SparseOpticalFlow
RyoheiHagimoto 0:0e0631af0305 597 {
RyoheiHagimoto 0:0e0631af0305 598 public:
RyoheiHagimoto 0:0e0631af0305 599 CV_WRAP virtual Size getWinSize() const = 0;
RyoheiHagimoto 0:0e0631af0305 600 CV_WRAP virtual void setWinSize(Size winSize) = 0;
RyoheiHagimoto 0:0e0631af0305 601
RyoheiHagimoto 0:0e0631af0305 602 CV_WRAP virtual int getMaxLevel() const = 0;
RyoheiHagimoto 0:0e0631af0305 603 CV_WRAP virtual void setMaxLevel(int maxLevel) = 0;
RyoheiHagimoto 0:0e0631af0305 604
RyoheiHagimoto 0:0e0631af0305 605 CV_WRAP virtual TermCriteria getTermCriteria() const = 0;
RyoheiHagimoto 0:0e0631af0305 606 CV_WRAP virtual void setTermCriteria(TermCriteria& crit) = 0;
RyoheiHagimoto 0:0e0631af0305 607
RyoheiHagimoto 0:0e0631af0305 608 CV_WRAP virtual int getFlags() const = 0;
RyoheiHagimoto 0:0e0631af0305 609 CV_WRAP virtual void setFlags(int flags) = 0;
RyoheiHagimoto 0:0e0631af0305 610
RyoheiHagimoto 0:0e0631af0305 611 CV_WRAP virtual double getMinEigThreshold() const = 0;
RyoheiHagimoto 0:0e0631af0305 612 CV_WRAP virtual void setMinEigThreshold(double minEigThreshold) = 0;
RyoheiHagimoto 0:0e0631af0305 613
RyoheiHagimoto 0:0e0631af0305 614 CV_WRAP static Ptr<SparsePyrLKOpticalFlow> create(
RyoheiHagimoto 0:0e0631af0305 615 Size winSize = Size(21, 21),
RyoheiHagimoto 0:0e0631af0305 616 int maxLevel = 3, TermCriteria crit =
RyoheiHagimoto 0:0e0631af0305 617 TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 30, 0.01),
RyoheiHagimoto 0:0e0631af0305 618 int flags = 0,
RyoheiHagimoto 0:0e0631af0305 619 double minEigThreshold = 1e-4);
RyoheiHagimoto 0:0e0631af0305 620 };
RyoheiHagimoto 0:0e0631af0305 621
RyoheiHagimoto 0:0e0631af0305 622 //! @} video_track
RyoheiHagimoto 0:0e0631af0305 623
RyoheiHagimoto 0:0e0631af0305 624 } // cv
RyoheiHagimoto 0:0e0631af0305 625
RyoheiHagimoto 0:0e0631af0305 626 #endif