Joe Verbout
/
main
opencv on mbed
opencv2/flann.hpp@0:ea44dc9ed014, 2016-03-31 (annotated)
- Committer:
- joeverbout
- Date:
- Thu Mar 31 21:16:38 2016 +0000
- Revision:
- 0:ea44dc9ed014
OpenCV on mbed attempt
Who changed what in which revision?
User | Revision | Line number | New contents of line |
---|---|---|---|
joeverbout | 0:ea44dc9ed014 | 1 | /*M/////////////////////////////////////////////////////////////////////////////////////// |
joeverbout | 0:ea44dc9ed014 | 2 | // |
joeverbout | 0:ea44dc9ed014 | 3 | // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING. |
joeverbout | 0:ea44dc9ed014 | 4 | // |
joeverbout | 0:ea44dc9ed014 | 5 | // By downloading, copying, installing or using the software you agree to this license. |
joeverbout | 0:ea44dc9ed014 | 6 | // If you do not agree to this license, do not download, install, |
joeverbout | 0:ea44dc9ed014 | 7 | // copy or use the software. |
joeverbout | 0:ea44dc9ed014 | 8 | // |
joeverbout | 0:ea44dc9ed014 | 9 | // |
joeverbout | 0:ea44dc9ed014 | 10 | // License Agreement |
joeverbout | 0:ea44dc9ed014 | 11 | // For Open Source Computer Vision Library |
joeverbout | 0:ea44dc9ed014 | 12 | // |
joeverbout | 0:ea44dc9ed014 | 13 | // Copyright (C) 2000-2008, Intel Corporation, all rights reserved. |
joeverbout | 0:ea44dc9ed014 | 14 | // Copyright (C) 2009, Willow Garage Inc., all rights reserved. |
joeverbout | 0:ea44dc9ed014 | 15 | // Third party copyrights are property of their respective owners. |
joeverbout | 0:ea44dc9ed014 | 16 | // |
joeverbout | 0:ea44dc9ed014 | 17 | // Redistribution and use in source and binary forms, with or without modification, |
joeverbout | 0:ea44dc9ed014 | 18 | // are permitted provided that the following conditions are met: |
joeverbout | 0:ea44dc9ed014 | 19 | // |
joeverbout | 0:ea44dc9ed014 | 20 | // * Redistribution's of source code must retain the above copyright notice, |
joeverbout | 0:ea44dc9ed014 | 21 | // this list of conditions and the following disclaimer. |
joeverbout | 0:ea44dc9ed014 | 22 | // |
joeverbout | 0:ea44dc9ed014 | 23 | // * Redistribution's in binary form must reproduce the above copyright notice, |
joeverbout | 0:ea44dc9ed014 | 24 | // this list of conditions and the following disclaimer in the documentation |
joeverbout | 0:ea44dc9ed014 | 25 | // and/or other materials provided with the distribution. |
joeverbout | 0:ea44dc9ed014 | 26 | // |
joeverbout | 0:ea44dc9ed014 | 27 | // * The name of the copyright holders may not be used to endorse or promote products |
joeverbout | 0:ea44dc9ed014 | 28 | // derived from this software without specific prior written permission. |
joeverbout | 0:ea44dc9ed014 | 29 | // |
joeverbout | 0:ea44dc9ed014 | 30 | // This software is provided by the copyright holders and contributors "as is" and |
joeverbout | 0:ea44dc9ed014 | 31 | // any express or implied warranties, including, but not limited to, the implied |
joeverbout | 0:ea44dc9ed014 | 32 | // warranties of merchantability and fitness for a particular purpose are disclaimed. |
joeverbout | 0:ea44dc9ed014 | 33 | // In no event shall the Intel Corporation or contributors be liable for any direct, |
joeverbout | 0:ea44dc9ed014 | 34 | // indirect, incidental, special, exemplary, or consequential damages |
joeverbout | 0:ea44dc9ed014 | 35 | // (including, but not limited to, procurement of substitute goods or services; |
joeverbout | 0:ea44dc9ed014 | 36 | // loss of use, data, or profits; or business interruption) however caused |
joeverbout | 0:ea44dc9ed014 | 37 | // and on any theory of liability, whether in contract, strict liability, |
joeverbout | 0:ea44dc9ed014 | 38 | // or tort (including negligence or otherwise) arising in any way out of |
joeverbout | 0:ea44dc9ed014 | 39 | // the use of this software, even if advised of the possibility of such damage. |
joeverbout | 0:ea44dc9ed014 | 40 | // |
joeverbout | 0:ea44dc9ed014 | 41 | //M*/ |
joeverbout | 0:ea44dc9ed014 | 42 | |
joeverbout | 0:ea44dc9ed014 | 43 | #ifndef _OPENCV_FLANN_HPP_ |
joeverbout | 0:ea44dc9ed014 | 44 | #define _OPENCV_FLANN_HPP_ |
joeverbout | 0:ea44dc9ed014 | 45 | |
joeverbout | 0:ea44dc9ed014 | 46 | #include "opencv2/core.hpp" |
joeverbout | 0:ea44dc9ed014 | 47 | #include "opencv2/flann/miniflann.hpp" |
joeverbout | 0:ea44dc9ed014 | 48 | #include "opencv2/flann/flann_base.hpp" |
joeverbout | 0:ea44dc9ed014 | 49 | |
joeverbout | 0:ea44dc9ed014 | 50 | /** |
joeverbout | 0:ea44dc9ed014 | 51 | @defgroup flann Clustering and Search in Multi-Dimensional Spaces |
joeverbout | 0:ea44dc9ed014 | 52 | |
joeverbout | 0:ea44dc9ed014 | 53 | This section documents OpenCV's interface to the FLANN library. FLANN (Fast Library for Approximate |
joeverbout | 0:ea44dc9ed014 | 54 | Nearest Neighbors) is a library that contains a collection of algorithms optimized for fast nearest |
joeverbout | 0:ea44dc9ed014 | 55 | neighbor search in large datasets and for high dimensional features. More information about FLANN |
joeverbout | 0:ea44dc9ed014 | 56 | can be found in @cite Muja2009 . |
joeverbout | 0:ea44dc9ed014 | 57 | */ |
joeverbout | 0:ea44dc9ed014 | 58 | |
joeverbout | 0:ea44dc9ed014 | 59 | namespace cvflann |
joeverbout | 0:ea44dc9ed014 | 60 | { |
joeverbout | 0:ea44dc9ed014 | 61 | CV_EXPORTS flann_distance_t flann_distance_type(); |
joeverbout | 0:ea44dc9ed014 | 62 | FLANN_DEPRECATED CV_EXPORTS void set_distance_type(flann_distance_t distance_type, int order); |
joeverbout | 0:ea44dc9ed014 | 63 | } |
joeverbout | 0:ea44dc9ed014 | 64 | |
joeverbout | 0:ea44dc9ed014 | 65 | |
joeverbout | 0:ea44dc9ed014 | 66 | namespace cv |
joeverbout | 0:ea44dc9ed014 | 67 | { |
joeverbout | 0:ea44dc9ed014 | 68 | namespace flann |
joeverbout | 0:ea44dc9ed014 | 69 | { |
joeverbout | 0:ea44dc9ed014 | 70 | |
joeverbout | 0:ea44dc9ed014 | 71 | |
joeverbout | 0:ea44dc9ed014 | 72 | //! @addtogroup flann |
joeverbout | 0:ea44dc9ed014 | 73 | //! @{ |
joeverbout | 0:ea44dc9ed014 | 74 | |
joeverbout | 0:ea44dc9ed014 | 75 | template <typename T> struct CvType {}; |
joeverbout | 0:ea44dc9ed014 | 76 | template <> struct CvType<unsigned char> { static int type() { return CV_8U; } }; |
joeverbout | 0:ea44dc9ed014 | 77 | template <> struct CvType<char> { static int type() { return CV_8S; } }; |
joeverbout | 0:ea44dc9ed014 | 78 | template <> struct CvType<unsigned short> { static int type() { return CV_16U; } }; |
joeverbout | 0:ea44dc9ed014 | 79 | template <> struct CvType<short> { static int type() { return CV_16S; } }; |
joeverbout | 0:ea44dc9ed014 | 80 | template <> struct CvType<int> { static int type() { return CV_32S; } }; |
joeverbout | 0:ea44dc9ed014 | 81 | template <> struct CvType<float> { static int type() { return CV_32F; } }; |
joeverbout | 0:ea44dc9ed014 | 82 | template <> struct CvType<double> { static int type() { return CV_64F; } }; |
joeverbout | 0:ea44dc9ed014 | 83 | |
joeverbout | 0:ea44dc9ed014 | 84 | |
joeverbout | 0:ea44dc9ed014 | 85 | // bring the flann parameters into this namespace |
joeverbout | 0:ea44dc9ed014 | 86 | using ::cvflann::get_param; |
joeverbout | 0:ea44dc9ed014 | 87 | using ::cvflann::print_params; |
joeverbout | 0:ea44dc9ed014 | 88 | |
joeverbout | 0:ea44dc9ed014 | 89 | // bring the flann distances into this namespace |
joeverbout | 0:ea44dc9ed014 | 90 | using ::cvflann::L2_Simple; |
joeverbout | 0:ea44dc9ed014 | 91 | using ::cvflann::L2; |
joeverbout | 0:ea44dc9ed014 | 92 | using ::cvflann::L1; |
joeverbout | 0:ea44dc9ed014 | 93 | using ::cvflann::MinkowskiDistance; |
joeverbout | 0:ea44dc9ed014 | 94 | using ::cvflann::MaxDistance; |
joeverbout | 0:ea44dc9ed014 | 95 | using ::cvflann::HammingLUT; |
joeverbout | 0:ea44dc9ed014 | 96 | using ::cvflann::Hamming; |
joeverbout | 0:ea44dc9ed014 | 97 | using ::cvflann::Hamming2; |
joeverbout | 0:ea44dc9ed014 | 98 | using ::cvflann::HistIntersectionDistance; |
joeverbout | 0:ea44dc9ed014 | 99 | using ::cvflann::HellingerDistance; |
joeverbout | 0:ea44dc9ed014 | 100 | using ::cvflann::ChiSquareDistance; |
joeverbout | 0:ea44dc9ed014 | 101 | using ::cvflann::KL_Divergence; |
joeverbout | 0:ea44dc9ed014 | 102 | |
joeverbout | 0:ea44dc9ed014 | 103 | |
joeverbout | 0:ea44dc9ed014 | 104 | /** @brief The FLANN nearest neighbor index class. This class is templated with the type of elements for which |
joeverbout | 0:ea44dc9ed014 | 105 | the index is built. |
joeverbout | 0:ea44dc9ed014 | 106 | */ |
joeverbout | 0:ea44dc9ed014 | 107 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 108 | class GenericIndex |
joeverbout | 0:ea44dc9ed014 | 109 | { |
joeverbout | 0:ea44dc9ed014 | 110 | public: |
joeverbout | 0:ea44dc9ed014 | 111 | typedef typename Distance::ElementType ElementType; |
joeverbout | 0:ea44dc9ed014 | 112 | typedef typename Distance::ResultType DistanceType; |
joeverbout | 0:ea44dc9ed014 | 113 | |
joeverbout | 0:ea44dc9ed014 | 114 | /** @brief Constructs a nearest neighbor search index for a given dataset. |
joeverbout | 0:ea44dc9ed014 | 115 | |
joeverbout | 0:ea44dc9ed014 | 116 | @param features Matrix of containing the features(points) to index. The size of the matrix is |
joeverbout | 0:ea44dc9ed014 | 117 | num_features x feature_dimensionality and the data type of the elements in the matrix must |
joeverbout | 0:ea44dc9ed014 | 118 | coincide with the type of the index. |
joeverbout | 0:ea44dc9ed014 | 119 | @param params Structure containing the index parameters. The type of index that will be |
joeverbout | 0:ea44dc9ed014 | 120 | constructed depends on the type of this parameter. See the description. |
joeverbout | 0:ea44dc9ed014 | 121 | @param distance |
joeverbout | 0:ea44dc9ed014 | 122 | |
joeverbout | 0:ea44dc9ed014 | 123 | The method constructs a fast search structure from a set of features using the specified algorithm |
joeverbout | 0:ea44dc9ed014 | 124 | with specified parameters, as defined by params. params is a reference to one of the following class |
joeverbout | 0:ea44dc9ed014 | 125 | IndexParams descendants: |
joeverbout | 0:ea44dc9ed014 | 126 | |
joeverbout | 0:ea44dc9ed014 | 127 | - **LinearIndexParams** When passing an object of this type, the index will perform a linear, |
joeverbout | 0:ea44dc9ed014 | 128 | brute-force search. : |
joeverbout | 0:ea44dc9ed014 | 129 | @code |
joeverbout | 0:ea44dc9ed014 | 130 | struct LinearIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 131 | { |
joeverbout | 0:ea44dc9ed014 | 132 | }; |
joeverbout | 0:ea44dc9ed014 | 133 | @endcode |
joeverbout | 0:ea44dc9ed014 | 134 | - **KDTreeIndexParams** When passing an object of this type the index constructed will consist of |
joeverbout | 0:ea44dc9ed014 | 135 | a set of randomized kd-trees which will be searched in parallel. : |
joeverbout | 0:ea44dc9ed014 | 136 | @code |
joeverbout | 0:ea44dc9ed014 | 137 | struct KDTreeIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 138 | { |
joeverbout | 0:ea44dc9ed014 | 139 | KDTreeIndexParams( int trees = 4 ); |
joeverbout | 0:ea44dc9ed014 | 140 | }; |
joeverbout | 0:ea44dc9ed014 | 141 | @endcode |
joeverbout | 0:ea44dc9ed014 | 142 | - **KMeansIndexParams** When passing an object of this type the index constructed will be a |
joeverbout | 0:ea44dc9ed014 | 143 | hierarchical k-means tree. : |
joeverbout | 0:ea44dc9ed014 | 144 | @code |
joeverbout | 0:ea44dc9ed014 | 145 | struct KMeansIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 146 | { |
joeverbout | 0:ea44dc9ed014 | 147 | KMeansIndexParams( |
joeverbout | 0:ea44dc9ed014 | 148 | int branching = 32, |
joeverbout | 0:ea44dc9ed014 | 149 | int iterations = 11, |
joeverbout | 0:ea44dc9ed014 | 150 | flann_centers_init_t centers_init = CENTERS_RANDOM, |
joeverbout | 0:ea44dc9ed014 | 151 | float cb_index = 0.2 ); |
joeverbout | 0:ea44dc9ed014 | 152 | }; |
joeverbout | 0:ea44dc9ed014 | 153 | @endcode |
joeverbout | 0:ea44dc9ed014 | 154 | - **CompositeIndexParams** When using a parameters object of this type the index created |
joeverbout | 0:ea44dc9ed014 | 155 | combines the randomized kd-trees and the hierarchical k-means tree. : |
joeverbout | 0:ea44dc9ed014 | 156 | @code |
joeverbout | 0:ea44dc9ed014 | 157 | struct CompositeIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 158 | { |
joeverbout | 0:ea44dc9ed014 | 159 | CompositeIndexParams( |
joeverbout | 0:ea44dc9ed014 | 160 | int trees = 4, |
joeverbout | 0:ea44dc9ed014 | 161 | int branching = 32, |
joeverbout | 0:ea44dc9ed014 | 162 | int iterations = 11, |
joeverbout | 0:ea44dc9ed014 | 163 | flann_centers_init_t centers_init = CENTERS_RANDOM, |
joeverbout | 0:ea44dc9ed014 | 164 | float cb_index = 0.2 ); |
joeverbout | 0:ea44dc9ed014 | 165 | }; |
joeverbout | 0:ea44dc9ed014 | 166 | @endcode |
joeverbout | 0:ea44dc9ed014 | 167 | - **LshIndexParams** When using a parameters object of this type the index created uses |
joeverbout | 0:ea44dc9ed014 | 168 | multi-probe LSH (by Multi-Probe LSH: Efficient Indexing for High-Dimensional Similarity Search |
joeverbout | 0:ea44dc9ed014 | 169 | by Qin Lv, William Josephson, Zhe Wang, Moses Charikar, Kai Li., Proceedings of the 33rd |
joeverbout | 0:ea44dc9ed014 | 170 | International Conference on Very Large Data Bases (VLDB). Vienna, Austria. September 2007) : |
joeverbout | 0:ea44dc9ed014 | 171 | @code |
joeverbout | 0:ea44dc9ed014 | 172 | struct LshIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 173 | { |
joeverbout | 0:ea44dc9ed014 | 174 | LshIndexParams( |
joeverbout | 0:ea44dc9ed014 | 175 | unsigned int table_number, |
joeverbout | 0:ea44dc9ed014 | 176 | unsigned int key_size, |
joeverbout | 0:ea44dc9ed014 | 177 | unsigned int multi_probe_level ); |
joeverbout | 0:ea44dc9ed014 | 178 | }; |
joeverbout | 0:ea44dc9ed014 | 179 | @endcode |
joeverbout | 0:ea44dc9ed014 | 180 | - **AutotunedIndexParams** When passing an object of this type the index created is |
joeverbout | 0:ea44dc9ed014 | 181 | automatically tuned to offer the best performance, by choosing the optimal index type |
joeverbout | 0:ea44dc9ed014 | 182 | (randomized kd-trees, hierarchical kmeans, linear) and parameters for the dataset provided. : |
joeverbout | 0:ea44dc9ed014 | 183 | @code |
joeverbout | 0:ea44dc9ed014 | 184 | struct AutotunedIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 185 | { |
joeverbout | 0:ea44dc9ed014 | 186 | AutotunedIndexParams( |
joeverbout | 0:ea44dc9ed014 | 187 | float target_precision = 0.9, |
joeverbout | 0:ea44dc9ed014 | 188 | float build_weight = 0.01, |
joeverbout | 0:ea44dc9ed014 | 189 | float memory_weight = 0, |
joeverbout | 0:ea44dc9ed014 | 190 | float sample_fraction = 0.1 ); |
joeverbout | 0:ea44dc9ed014 | 191 | }; |
joeverbout | 0:ea44dc9ed014 | 192 | @endcode |
joeverbout | 0:ea44dc9ed014 | 193 | - **SavedIndexParams** This object type is used for loading a previously saved index from the |
joeverbout | 0:ea44dc9ed014 | 194 | disk. : |
joeverbout | 0:ea44dc9ed014 | 195 | @code |
joeverbout | 0:ea44dc9ed014 | 196 | struct SavedIndexParams : public IndexParams |
joeverbout | 0:ea44dc9ed014 | 197 | { |
joeverbout | 0:ea44dc9ed014 | 198 | SavedIndexParams( String filename ); |
joeverbout | 0:ea44dc9ed014 | 199 | }; |
joeverbout | 0:ea44dc9ed014 | 200 | @endcode |
joeverbout | 0:ea44dc9ed014 | 201 | */ |
joeverbout | 0:ea44dc9ed014 | 202 | GenericIndex(const Mat& features, const ::cvflann::IndexParams& params, Distance distance = Distance()); |
joeverbout | 0:ea44dc9ed014 | 203 | |
joeverbout | 0:ea44dc9ed014 | 204 | ~GenericIndex(); |
joeverbout | 0:ea44dc9ed014 | 205 | |
joeverbout | 0:ea44dc9ed014 | 206 | /** @brief Performs a K-nearest neighbor search for a given query point using the index. |
joeverbout | 0:ea44dc9ed014 | 207 | |
joeverbout | 0:ea44dc9ed014 | 208 | @param query The query point |
joeverbout | 0:ea44dc9ed014 | 209 | @param indices Vector that will contain the indices of the K-nearest neighbors found. It must have |
joeverbout | 0:ea44dc9ed014 | 210 | at least knn size. |
joeverbout | 0:ea44dc9ed014 | 211 | @param dists Vector that will contain the distances to the K-nearest neighbors found. It must have |
joeverbout | 0:ea44dc9ed014 | 212 | at least knn size. |
joeverbout | 0:ea44dc9ed014 | 213 | @param knn Number of nearest neighbors to search for. |
joeverbout | 0:ea44dc9ed014 | 214 | @param params SearchParams |
joeverbout | 0:ea44dc9ed014 | 215 | */ |
joeverbout | 0:ea44dc9ed014 | 216 | void knnSearch(const std::vector<ElementType>& query, std::vector<int>& indices, |
joeverbout | 0:ea44dc9ed014 | 217 | std::vector<DistanceType>& dists, int knn, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 218 | void knnSearch(const Mat& queries, Mat& indices, Mat& dists, int knn, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 219 | |
joeverbout | 0:ea44dc9ed014 | 220 | int radiusSearch(const std::vector<ElementType>& query, std::vector<int>& indices, |
joeverbout | 0:ea44dc9ed014 | 221 | std::vector<DistanceType>& dists, DistanceType radius, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 222 | int radiusSearch(const Mat& query, Mat& indices, Mat& dists, |
joeverbout | 0:ea44dc9ed014 | 223 | DistanceType radius, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 224 | |
joeverbout | 0:ea44dc9ed014 | 225 | void save(String filename) { nnIndex->save(filename); } |
joeverbout | 0:ea44dc9ed014 | 226 | |
joeverbout | 0:ea44dc9ed014 | 227 | int veclen() const { return nnIndex->veclen(); } |
joeverbout | 0:ea44dc9ed014 | 228 | |
joeverbout | 0:ea44dc9ed014 | 229 | int size() const { return nnIndex->size(); } |
joeverbout | 0:ea44dc9ed014 | 230 | |
joeverbout | 0:ea44dc9ed014 | 231 | ::cvflann::IndexParams getParameters() { return nnIndex->getParameters(); } |
joeverbout | 0:ea44dc9ed014 | 232 | |
joeverbout | 0:ea44dc9ed014 | 233 | FLANN_DEPRECATED const ::cvflann::IndexParams* getIndexParameters() { return nnIndex->getIndexParameters(); } |
joeverbout | 0:ea44dc9ed014 | 234 | |
joeverbout | 0:ea44dc9ed014 | 235 | private: |
joeverbout | 0:ea44dc9ed014 | 236 | ::cvflann::Index<Distance>* nnIndex; |
joeverbout | 0:ea44dc9ed014 | 237 | }; |
joeverbout | 0:ea44dc9ed014 | 238 | |
joeverbout | 0:ea44dc9ed014 | 239 | //! @cond IGNORED |
joeverbout | 0:ea44dc9ed014 | 240 | |
joeverbout | 0:ea44dc9ed014 | 241 | #define FLANN_DISTANCE_CHECK \ |
joeverbout | 0:ea44dc9ed014 | 242 | if ( ::cvflann::flann_distance_type() != cvflann::FLANN_DIST_L2) { \ |
joeverbout | 0:ea44dc9ed014 | 243 | printf("[WARNING] You are using cv::flann::Index (or cv::flann::GenericIndex) and have also changed "\ |
joeverbout | 0:ea44dc9ed014 | 244 | "the distance using cvflann::set_distance_type. This is no longer working as expected "\ |
joeverbout | 0:ea44dc9ed014 | 245 | "(cv::flann::Index always uses L2). You should create the index templated on the distance, "\ |
joeverbout | 0:ea44dc9ed014 | 246 | "for example for L1 distance use: GenericIndex< L1<float> > \n"); \ |
joeverbout | 0:ea44dc9ed014 | 247 | } |
joeverbout | 0:ea44dc9ed014 | 248 | |
joeverbout | 0:ea44dc9ed014 | 249 | |
joeverbout | 0:ea44dc9ed014 | 250 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 251 | GenericIndex<Distance>::GenericIndex(const Mat& dataset, const ::cvflann::IndexParams& params, Distance distance) |
joeverbout | 0:ea44dc9ed014 | 252 | { |
joeverbout | 0:ea44dc9ed014 | 253 | CV_Assert(dataset.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 254 | CV_Assert(dataset.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 255 | ::cvflann::Matrix<ElementType> m_dataset((ElementType*)dataset.ptr<ElementType>(0), dataset.rows, dataset.cols); |
joeverbout | 0:ea44dc9ed014 | 256 | |
joeverbout | 0:ea44dc9ed014 | 257 | nnIndex = new ::cvflann::Index<Distance>(m_dataset, params, distance); |
joeverbout | 0:ea44dc9ed014 | 258 | |
joeverbout | 0:ea44dc9ed014 | 259 | FLANN_DISTANCE_CHECK |
joeverbout | 0:ea44dc9ed014 | 260 | |
joeverbout | 0:ea44dc9ed014 | 261 | nnIndex->buildIndex(); |
joeverbout | 0:ea44dc9ed014 | 262 | } |
joeverbout | 0:ea44dc9ed014 | 263 | |
joeverbout | 0:ea44dc9ed014 | 264 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 265 | GenericIndex<Distance>::~GenericIndex() |
joeverbout | 0:ea44dc9ed014 | 266 | { |
joeverbout | 0:ea44dc9ed014 | 267 | delete nnIndex; |
joeverbout | 0:ea44dc9ed014 | 268 | } |
joeverbout | 0:ea44dc9ed014 | 269 | |
joeverbout | 0:ea44dc9ed014 | 270 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 271 | void GenericIndex<Distance>::knnSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, int knn, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 272 | { |
joeverbout | 0:ea44dc9ed014 | 273 | ::cvflann::Matrix<ElementType> m_query((ElementType*)&query[0], 1, query.size()); |
joeverbout | 0:ea44dc9ed014 | 274 | ::cvflann::Matrix<int> m_indices(&indices[0], 1, indices.size()); |
joeverbout | 0:ea44dc9ed014 | 275 | ::cvflann::Matrix<DistanceType> m_dists(&dists[0], 1, dists.size()); |
joeverbout | 0:ea44dc9ed014 | 276 | |
joeverbout | 0:ea44dc9ed014 | 277 | FLANN_DISTANCE_CHECK |
joeverbout | 0:ea44dc9ed014 | 278 | |
joeverbout | 0:ea44dc9ed014 | 279 | nnIndex->knnSearch(m_query,m_indices,m_dists,knn,searchParams); |
joeverbout | 0:ea44dc9ed014 | 280 | } |
joeverbout | 0:ea44dc9ed014 | 281 | |
joeverbout | 0:ea44dc9ed014 | 282 | |
joeverbout | 0:ea44dc9ed014 | 283 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 284 | void GenericIndex<Distance>::knnSearch(const Mat& queries, Mat& indices, Mat& dists, int knn, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 285 | { |
joeverbout | 0:ea44dc9ed014 | 286 | CV_Assert(queries.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 287 | CV_Assert(queries.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 288 | ::cvflann::Matrix<ElementType> m_queries((ElementType*)queries.ptr<ElementType>(0), queries.rows, queries.cols); |
joeverbout | 0:ea44dc9ed014 | 289 | |
joeverbout | 0:ea44dc9ed014 | 290 | CV_Assert(indices.type() == CV_32S); |
joeverbout | 0:ea44dc9ed014 | 291 | CV_Assert(indices.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 292 | ::cvflann::Matrix<int> m_indices((int*)indices.ptr<int>(0), indices.rows, indices.cols); |
joeverbout | 0:ea44dc9ed014 | 293 | |
joeverbout | 0:ea44dc9ed014 | 294 | CV_Assert(dists.type() == CvType<DistanceType>::type()); |
joeverbout | 0:ea44dc9ed014 | 295 | CV_Assert(dists.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 296 | ::cvflann::Matrix<DistanceType> m_dists((DistanceType*)dists.ptr<DistanceType>(0), dists.rows, dists.cols); |
joeverbout | 0:ea44dc9ed014 | 297 | |
joeverbout | 0:ea44dc9ed014 | 298 | FLANN_DISTANCE_CHECK |
joeverbout | 0:ea44dc9ed014 | 299 | |
joeverbout | 0:ea44dc9ed014 | 300 | nnIndex->knnSearch(m_queries,m_indices,m_dists,knn, searchParams); |
joeverbout | 0:ea44dc9ed014 | 301 | } |
joeverbout | 0:ea44dc9ed014 | 302 | |
joeverbout | 0:ea44dc9ed014 | 303 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 304 | int GenericIndex<Distance>::radiusSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, DistanceType radius, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 305 | { |
joeverbout | 0:ea44dc9ed014 | 306 | ::cvflann::Matrix<ElementType> m_query((ElementType*)&query[0], 1, query.size()); |
joeverbout | 0:ea44dc9ed014 | 307 | ::cvflann::Matrix<int> m_indices(&indices[0], 1, indices.size()); |
joeverbout | 0:ea44dc9ed014 | 308 | ::cvflann::Matrix<DistanceType> m_dists(&dists[0], 1, dists.size()); |
joeverbout | 0:ea44dc9ed014 | 309 | |
joeverbout | 0:ea44dc9ed014 | 310 | FLANN_DISTANCE_CHECK |
joeverbout | 0:ea44dc9ed014 | 311 | |
joeverbout | 0:ea44dc9ed014 | 312 | return nnIndex->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 313 | } |
joeverbout | 0:ea44dc9ed014 | 314 | |
joeverbout | 0:ea44dc9ed014 | 315 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 316 | int GenericIndex<Distance>::radiusSearch(const Mat& query, Mat& indices, Mat& dists, DistanceType radius, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 317 | { |
joeverbout | 0:ea44dc9ed014 | 318 | CV_Assert(query.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 319 | CV_Assert(query.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 320 | ::cvflann::Matrix<ElementType> m_query((ElementType*)query.ptr<ElementType>(0), query.rows, query.cols); |
joeverbout | 0:ea44dc9ed014 | 321 | |
joeverbout | 0:ea44dc9ed014 | 322 | CV_Assert(indices.type() == CV_32S); |
joeverbout | 0:ea44dc9ed014 | 323 | CV_Assert(indices.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 324 | ::cvflann::Matrix<int> m_indices((int*)indices.ptr<int>(0), indices.rows, indices.cols); |
joeverbout | 0:ea44dc9ed014 | 325 | |
joeverbout | 0:ea44dc9ed014 | 326 | CV_Assert(dists.type() == CvType<DistanceType>::type()); |
joeverbout | 0:ea44dc9ed014 | 327 | CV_Assert(dists.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 328 | ::cvflann::Matrix<DistanceType> m_dists((DistanceType*)dists.ptr<DistanceType>(0), dists.rows, dists.cols); |
joeverbout | 0:ea44dc9ed014 | 329 | |
joeverbout | 0:ea44dc9ed014 | 330 | FLANN_DISTANCE_CHECK |
joeverbout | 0:ea44dc9ed014 | 331 | |
joeverbout | 0:ea44dc9ed014 | 332 | return nnIndex->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 333 | } |
joeverbout | 0:ea44dc9ed014 | 334 | |
joeverbout | 0:ea44dc9ed014 | 335 | //! @endcond |
joeverbout | 0:ea44dc9ed014 | 336 | |
joeverbout | 0:ea44dc9ed014 | 337 | /** |
joeverbout | 0:ea44dc9ed014 | 338 | * @deprecated Use GenericIndex class instead |
joeverbout | 0:ea44dc9ed014 | 339 | */ |
joeverbout | 0:ea44dc9ed014 | 340 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 341 | class |
joeverbout | 0:ea44dc9ed014 | 342 | #ifndef _MSC_VER |
joeverbout | 0:ea44dc9ed014 | 343 | FLANN_DEPRECATED |
joeverbout | 0:ea44dc9ed014 | 344 | #endif |
joeverbout | 0:ea44dc9ed014 | 345 | Index_ { |
joeverbout | 0:ea44dc9ed014 | 346 | public: |
joeverbout | 0:ea44dc9ed014 | 347 | typedef typename L2<T>::ElementType ElementType; |
joeverbout | 0:ea44dc9ed014 | 348 | typedef typename L2<T>::ResultType DistanceType; |
joeverbout | 0:ea44dc9ed014 | 349 | |
joeverbout | 0:ea44dc9ed014 | 350 | Index_(const Mat& features, const ::cvflann::IndexParams& params); |
joeverbout | 0:ea44dc9ed014 | 351 | |
joeverbout | 0:ea44dc9ed014 | 352 | ~Index_(); |
joeverbout | 0:ea44dc9ed014 | 353 | |
joeverbout | 0:ea44dc9ed014 | 354 | void knnSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, int knn, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 355 | void knnSearch(const Mat& queries, Mat& indices, Mat& dists, int knn, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 356 | |
joeverbout | 0:ea44dc9ed014 | 357 | int radiusSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, DistanceType radius, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 358 | int radiusSearch(const Mat& query, Mat& indices, Mat& dists, DistanceType radius, const ::cvflann::SearchParams& params); |
joeverbout | 0:ea44dc9ed014 | 359 | |
joeverbout | 0:ea44dc9ed014 | 360 | void save(String filename) |
joeverbout | 0:ea44dc9ed014 | 361 | { |
joeverbout | 0:ea44dc9ed014 | 362 | if (nnIndex_L1) nnIndex_L1->save(filename); |
joeverbout | 0:ea44dc9ed014 | 363 | if (nnIndex_L2) nnIndex_L2->save(filename); |
joeverbout | 0:ea44dc9ed014 | 364 | } |
joeverbout | 0:ea44dc9ed014 | 365 | |
joeverbout | 0:ea44dc9ed014 | 366 | int veclen() const |
joeverbout | 0:ea44dc9ed014 | 367 | { |
joeverbout | 0:ea44dc9ed014 | 368 | if (nnIndex_L1) return nnIndex_L1->veclen(); |
joeverbout | 0:ea44dc9ed014 | 369 | if (nnIndex_L2) return nnIndex_L2->veclen(); |
joeverbout | 0:ea44dc9ed014 | 370 | } |
joeverbout | 0:ea44dc9ed014 | 371 | |
joeverbout | 0:ea44dc9ed014 | 372 | int size() const |
joeverbout | 0:ea44dc9ed014 | 373 | { |
joeverbout | 0:ea44dc9ed014 | 374 | if (nnIndex_L1) return nnIndex_L1->size(); |
joeverbout | 0:ea44dc9ed014 | 375 | if (nnIndex_L2) return nnIndex_L2->size(); |
joeverbout | 0:ea44dc9ed014 | 376 | } |
joeverbout | 0:ea44dc9ed014 | 377 | |
joeverbout | 0:ea44dc9ed014 | 378 | ::cvflann::IndexParams getParameters() |
joeverbout | 0:ea44dc9ed014 | 379 | { |
joeverbout | 0:ea44dc9ed014 | 380 | if (nnIndex_L1) return nnIndex_L1->getParameters(); |
joeverbout | 0:ea44dc9ed014 | 381 | if (nnIndex_L2) return nnIndex_L2->getParameters(); |
joeverbout | 0:ea44dc9ed014 | 382 | |
joeverbout | 0:ea44dc9ed014 | 383 | } |
joeverbout | 0:ea44dc9ed014 | 384 | |
joeverbout | 0:ea44dc9ed014 | 385 | FLANN_DEPRECATED const ::cvflann::IndexParams* getIndexParameters() |
joeverbout | 0:ea44dc9ed014 | 386 | { |
joeverbout | 0:ea44dc9ed014 | 387 | if (nnIndex_L1) return nnIndex_L1->getIndexParameters(); |
joeverbout | 0:ea44dc9ed014 | 388 | if (nnIndex_L2) return nnIndex_L2->getIndexParameters(); |
joeverbout | 0:ea44dc9ed014 | 389 | } |
joeverbout | 0:ea44dc9ed014 | 390 | |
joeverbout | 0:ea44dc9ed014 | 391 | private: |
joeverbout | 0:ea44dc9ed014 | 392 | // providing backwards compatibility for L2 and L1 distances (most common) |
joeverbout | 0:ea44dc9ed014 | 393 | ::cvflann::Index< L2<ElementType> >* nnIndex_L2; |
joeverbout | 0:ea44dc9ed014 | 394 | ::cvflann::Index< L1<ElementType> >* nnIndex_L1; |
joeverbout | 0:ea44dc9ed014 | 395 | }; |
joeverbout | 0:ea44dc9ed014 | 396 | |
joeverbout | 0:ea44dc9ed014 | 397 | #ifdef _MSC_VER |
joeverbout | 0:ea44dc9ed014 | 398 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 399 | class FLANN_DEPRECATED Index_; |
joeverbout | 0:ea44dc9ed014 | 400 | #endif |
joeverbout | 0:ea44dc9ed014 | 401 | |
joeverbout | 0:ea44dc9ed014 | 402 | //! @cond IGNORED |
joeverbout | 0:ea44dc9ed014 | 403 | |
joeverbout | 0:ea44dc9ed014 | 404 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 405 | Index_<T>::Index_(const Mat& dataset, const ::cvflann::IndexParams& params) |
joeverbout | 0:ea44dc9ed014 | 406 | { |
joeverbout | 0:ea44dc9ed014 | 407 | printf("[WARNING] The cv::flann::Index_<T> class is deperecated, use cv::flann::GenericIndex<Distance> instead\n"); |
joeverbout | 0:ea44dc9ed014 | 408 | |
joeverbout | 0:ea44dc9ed014 | 409 | CV_Assert(dataset.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 410 | CV_Assert(dataset.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 411 | ::cvflann::Matrix<ElementType> m_dataset((ElementType*)dataset.ptr<ElementType>(0), dataset.rows, dataset.cols); |
joeverbout | 0:ea44dc9ed014 | 412 | |
joeverbout | 0:ea44dc9ed014 | 413 | if ( ::cvflann::flann_distance_type() == cvflann::FLANN_DIST_L2 ) { |
joeverbout | 0:ea44dc9ed014 | 414 | nnIndex_L1 = NULL; |
joeverbout | 0:ea44dc9ed014 | 415 | nnIndex_L2 = new ::cvflann::Index< L2<ElementType> >(m_dataset, params); |
joeverbout | 0:ea44dc9ed014 | 416 | } |
joeverbout | 0:ea44dc9ed014 | 417 | else if ( ::cvflann::flann_distance_type() == cvflann::FLANN_DIST_L1 ) { |
joeverbout | 0:ea44dc9ed014 | 418 | nnIndex_L1 = new ::cvflann::Index< L1<ElementType> >(m_dataset, params); |
joeverbout | 0:ea44dc9ed014 | 419 | nnIndex_L2 = NULL; |
joeverbout | 0:ea44dc9ed014 | 420 | } |
joeverbout | 0:ea44dc9ed014 | 421 | else { |
joeverbout | 0:ea44dc9ed014 | 422 | printf("[ERROR] cv::flann::Index_<T> only provides backwards compatibility for the L1 and L2 distances. " |
joeverbout | 0:ea44dc9ed014 | 423 | "For other distance types you must use cv::flann::GenericIndex<Distance>\n"); |
joeverbout | 0:ea44dc9ed014 | 424 | CV_Assert(0); |
joeverbout | 0:ea44dc9ed014 | 425 | } |
joeverbout | 0:ea44dc9ed014 | 426 | if (nnIndex_L1) nnIndex_L1->buildIndex(); |
joeverbout | 0:ea44dc9ed014 | 427 | if (nnIndex_L2) nnIndex_L2->buildIndex(); |
joeverbout | 0:ea44dc9ed014 | 428 | } |
joeverbout | 0:ea44dc9ed014 | 429 | |
joeverbout | 0:ea44dc9ed014 | 430 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 431 | Index_<T>::~Index_() |
joeverbout | 0:ea44dc9ed014 | 432 | { |
joeverbout | 0:ea44dc9ed014 | 433 | if (nnIndex_L1) delete nnIndex_L1; |
joeverbout | 0:ea44dc9ed014 | 434 | if (nnIndex_L2) delete nnIndex_L2; |
joeverbout | 0:ea44dc9ed014 | 435 | } |
joeverbout | 0:ea44dc9ed014 | 436 | |
joeverbout | 0:ea44dc9ed014 | 437 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 438 | void Index_<T>::knnSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, int knn, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 439 | { |
joeverbout | 0:ea44dc9ed014 | 440 | ::cvflann::Matrix<ElementType> m_query((ElementType*)&query[0], 1, query.size()); |
joeverbout | 0:ea44dc9ed014 | 441 | ::cvflann::Matrix<int> m_indices(&indices[0], 1, indices.size()); |
joeverbout | 0:ea44dc9ed014 | 442 | ::cvflann::Matrix<DistanceType> m_dists(&dists[0], 1, dists.size()); |
joeverbout | 0:ea44dc9ed014 | 443 | |
joeverbout | 0:ea44dc9ed014 | 444 | if (nnIndex_L1) nnIndex_L1->knnSearch(m_query,m_indices,m_dists,knn,searchParams); |
joeverbout | 0:ea44dc9ed014 | 445 | if (nnIndex_L2) nnIndex_L2->knnSearch(m_query,m_indices,m_dists,knn,searchParams); |
joeverbout | 0:ea44dc9ed014 | 446 | } |
joeverbout | 0:ea44dc9ed014 | 447 | |
joeverbout | 0:ea44dc9ed014 | 448 | |
joeverbout | 0:ea44dc9ed014 | 449 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 450 | void Index_<T>::knnSearch(const Mat& queries, Mat& indices, Mat& dists, int knn, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 451 | { |
joeverbout | 0:ea44dc9ed014 | 452 | CV_Assert(queries.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 453 | CV_Assert(queries.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 454 | ::cvflann::Matrix<ElementType> m_queries((ElementType*)queries.ptr<ElementType>(0), queries.rows, queries.cols); |
joeverbout | 0:ea44dc9ed014 | 455 | |
joeverbout | 0:ea44dc9ed014 | 456 | CV_Assert(indices.type() == CV_32S); |
joeverbout | 0:ea44dc9ed014 | 457 | CV_Assert(indices.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 458 | ::cvflann::Matrix<int> m_indices((int*)indices.ptr<int>(0), indices.rows, indices.cols); |
joeverbout | 0:ea44dc9ed014 | 459 | |
joeverbout | 0:ea44dc9ed014 | 460 | CV_Assert(dists.type() == CvType<DistanceType>::type()); |
joeverbout | 0:ea44dc9ed014 | 461 | CV_Assert(dists.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 462 | ::cvflann::Matrix<DistanceType> m_dists((DistanceType*)dists.ptr<DistanceType>(0), dists.rows, dists.cols); |
joeverbout | 0:ea44dc9ed014 | 463 | |
joeverbout | 0:ea44dc9ed014 | 464 | if (nnIndex_L1) nnIndex_L1->knnSearch(m_queries,m_indices,m_dists,knn, searchParams); |
joeverbout | 0:ea44dc9ed014 | 465 | if (nnIndex_L2) nnIndex_L2->knnSearch(m_queries,m_indices,m_dists,knn, searchParams); |
joeverbout | 0:ea44dc9ed014 | 466 | } |
joeverbout | 0:ea44dc9ed014 | 467 | |
joeverbout | 0:ea44dc9ed014 | 468 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 469 | int Index_<T>::radiusSearch(const std::vector<ElementType>& query, std::vector<int>& indices, std::vector<DistanceType>& dists, DistanceType radius, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 470 | { |
joeverbout | 0:ea44dc9ed014 | 471 | ::cvflann::Matrix<ElementType> m_query((ElementType*)&query[0], 1, query.size()); |
joeverbout | 0:ea44dc9ed014 | 472 | ::cvflann::Matrix<int> m_indices(&indices[0], 1, indices.size()); |
joeverbout | 0:ea44dc9ed014 | 473 | ::cvflann::Matrix<DistanceType> m_dists(&dists[0], 1, dists.size()); |
joeverbout | 0:ea44dc9ed014 | 474 | |
joeverbout | 0:ea44dc9ed014 | 475 | if (nnIndex_L1) return nnIndex_L1->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 476 | if (nnIndex_L2) return nnIndex_L2->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 477 | } |
joeverbout | 0:ea44dc9ed014 | 478 | |
joeverbout | 0:ea44dc9ed014 | 479 | template <typename T> |
joeverbout | 0:ea44dc9ed014 | 480 | int Index_<T>::radiusSearch(const Mat& query, Mat& indices, Mat& dists, DistanceType radius, const ::cvflann::SearchParams& searchParams) |
joeverbout | 0:ea44dc9ed014 | 481 | { |
joeverbout | 0:ea44dc9ed014 | 482 | CV_Assert(query.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 483 | CV_Assert(query.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 484 | ::cvflann::Matrix<ElementType> m_query((ElementType*)query.ptr<ElementType>(0), query.rows, query.cols); |
joeverbout | 0:ea44dc9ed014 | 485 | |
joeverbout | 0:ea44dc9ed014 | 486 | CV_Assert(indices.type() == CV_32S); |
joeverbout | 0:ea44dc9ed014 | 487 | CV_Assert(indices.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 488 | ::cvflann::Matrix<int> m_indices((int*)indices.ptr<int>(0), indices.rows, indices.cols); |
joeverbout | 0:ea44dc9ed014 | 489 | |
joeverbout | 0:ea44dc9ed014 | 490 | CV_Assert(dists.type() == CvType<DistanceType>::type()); |
joeverbout | 0:ea44dc9ed014 | 491 | CV_Assert(dists.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 492 | ::cvflann::Matrix<DistanceType> m_dists((DistanceType*)dists.ptr<DistanceType>(0), dists.rows, dists.cols); |
joeverbout | 0:ea44dc9ed014 | 493 | |
joeverbout | 0:ea44dc9ed014 | 494 | if (nnIndex_L1) return nnIndex_L1->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 495 | if (nnIndex_L2) return nnIndex_L2->radiusSearch(m_query,m_indices,m_dists,radius,searchParams); |
joeverbout | 0:ea44dc9ed014 | 496 | } |
joeverbout | 0:ea44dc9ed014 | 497 | |
joeverbout | 0:ea44dc9ed014 | 498 | //! @endcond |
joeverbout | 0:ea44dc9ed014 | 499 | |
joeverbout | 0:ea44dc9ed014 | 500 | /** @brief Clusters features using hierarchical k-means algorithm. |
joeverbout | 0:ea44dc9ed014 | 501 | |
joeverbout | 0:ea44dc9ed014 | 502 | @param features The points to be clustered. The matrix must have elements of type |
joeverbout | 0:ea44dc9ed014 | 503 | Distance::ElementType. |
joeverbout | 0:ea44dc9ed014 | 504 | @param centers The centers of the clusters obtained. The matrix must have type |
joeverbout | 0:ea44dc9ed014 | 505 | Distance::ResultType. The number of rows in this matrix represents the number of clusters desired, |
joeverbout | 0:ea44dc9ed014 | 506 | however, because of the way the cut in the hierarchical tree is chosen, the number of clusters |
joeverbout | 0:ea44dc9ed014 | 507 | computed will be the highest number of the form (branching-1)\*k+1 that's lower than the number of |
joeverbout | 0:ea44dc9ed014 | 508 | clusters desired, where branching is the tree's branching factor (see description of the |
joeverbout | 0:ea44dc9ed014 | 509 | KMeansIndexParams). |
joeverbout | 0:ea44dc9ed014 | 510 | @param params Parameters used in the construction of the hierarchical k-means tree. |
joeverbout | 0:ea44dc9ed014 | 511 | @param d Distance to be used for clustering. |
joeverbout | 0:ea44dc9ed014 | 512 | |
joeverbout | 0:ea44dc9ed014 | 513 | The method clusters the given feature vectors by constructing a hierarchical k-means tree and |
joeverbout | 0:ea44dc9ed014 | 514 | choosing a cut in the tree that minimizes the cluster's variance. It returns the number of clusters |
joeverbout | 0:ea44dc9ed014 | 515 | found. |
joeverbout | 0:ea44dc9ed014 | 516 | */ |
joeverbout | 0:ea44dc9ed014 | 517 | template <typename Distance> |
joeverbout | 0:ea44dc9ed014 | 518 | int hierarchicalClustering(const Mat& features, Mat& centers, const ::cvflann::KMeansIndexParams& params, |
joeverbout | 0:ea44dc9ed014 | 519 | Distance d = Distance()) |
joeverbout | 0:ea44dc9ed014 | 520 | { |
joeverbout | 0:ea44dc9ed014 | 521 | typedef typename Distance::ElementType ElementType; |
joeverbout | 0:ea44dc9ed014 | 522 | typedef typename Distance::ResultType DistanceType; |
joeverbout | 0:ea44dc9ed014 | 523 | |
joeverbout | 0:ea44dc9ed014 | 524 | CV_Assert(features.type() == CvType<ElementType>::type()); |
joeverbout | 0:ea44dc9ed014 | 525 | CV_Assert(features.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 526 | ::cvflann::Matrix<ElementType> m_features((ElementType*)features.ptr<ElementType>(0), features.rows, features.cols); |
joeverbout | 0:ea44dc9ed014 | 527 | |
joeverbout | 0:ea44dc9ed014 | 528 | CV_Assert(centers.type() == CvType<DistanceType>::type()); |
joeverbout | 0:ea44dc9ed014 | 529 | CV_Assert(centers.isContinuous()); |
joeverbout | 0:ea44dc9ed014 | 530 | ::cvflann::Matrix<DistanceType> m_centers((DistanceType*)centers.ptr<DistanceType>(0), centers.rows, centers.cols); |
joeverbout | 0:ea44dc9ed014 | 531 | |
joeverbout | 0:ea44dc9ed014 | 532 | return ::cvflann::hierarchicalClustering<Distance>(m_features, m_centers, params, d); |
joeverbout | 0:ea44dc9ed014 | 533 | } |
joeverbout | 0:ea44dc9ed014 | 534 | |
joeverbout | 0:ea44dc9ed014 | 535 | /** @deprecated |
joeverbout | 0:ea44dc9ed014 | 536 | */ |
joeverbout | 0:ea44dc9ed014 | 537 | template <typename ELEM_TYPE, typename DIST_TYPE> |
joeverbout | 0:ea44dc9ed014 | 538 | FLANN_DEPRECATED int hierarchicalClustering(const Mat& features, Mat& centers, const ::cvflann::KMeansIndexParams& params) |
joeverbout | 0:ea44dc9ed014 | 539 | { |
joeverbout | 0:ea44dc9ed014 | 540 | printf("[WARNING] cv::flann::hierarchicalClustering<ELEM_TYPE,DIST_TYPE> is deprecated, use " |
joeverbout | 0:ea44dc9ed014 | 541 | "cv::flann::hierarchicalClustering<Distance> instead\n"); |
joeverbout | 0:ea44dc9ed014 | 542 | |
joeverbout | 0:ea44dc9ed014 | 543 | if ( ::cvflann::flann_distance_type() == cvflann::FLANN_DIST_L2 ) { |
joeverbout | 0:ea44dc9ed014 | 544 | return hierarchicalClustering< L2<ELEM_TYPE> >(features, centers, params); |
joeverbout | 0:ea44dc9ed014 | 545 | } |
joeverbout | 0:ea44dc9ed014 | 546 | else if ( ::cvflann::flann_distance_type() == cvflann::FLANN_DIST_L1 ) { |
joeverbout | 0:ea44dc9ed014 | 547 | return hierarchicalClustering< L1<ELEM_TYPE> >(features, centers, params); |
joeverbout | 0:ea44dc9ed014 | 548 | } |
joeverbout | 0:ea44dc9ed014 | 549 | else { |
joeverbout | 0:ea44dc9ed014 | 550 | printf("[ERROR] cv::flann::hierarchicalClustering<ELEM_TYPE,DIST_TYPE> only provides backwards " |
joeverbout | 0:ea44dc9ed014 | 551 | "compatibility for the L1 and L2 distances. " |
joeverbout | 0:ea44dc9ed014 | 552 | "For other distance types you must use cv::flann::hierarchicalClustering<Distance>\n"); |
joeverbout | 0:ea44dc9ed014 | 553 | CV_Assert(0); |
joeverbout | 0:ea44dc9ed014 | 554 | } |
joeverbout | 0:ea44dc9ed014 | 555 | } |
joeverbout | 0:ea44dc9ed014 | 556 | |
joeverbout | 0:ea44dc9ed014 | 557 | //! @} flann |
joeverbout | 0:ea44dc9ed014 | 558 | |
joeverbout | 0:ea44dc9ed014 | 559 | } } // namespace cv::flann |
joeverbout | 0:ea44dc9ed014 | 560 | |
joeverbout | 0:ea44dc9ed014 | 561 | #endif |
joeverbout | 0:ea44dc9ed014 | 562 |