1 /*M///////////////////////////////////////////////////////////////////////////////////////
3 // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
5 // By downloading, copying, installing or using the software you agree to this license.
6 // If you do not agree to this license, do not download, install,
7 // copy or use the software.
11 // For Open Source Computer Vision Library
13 // Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
14 // Copyright (C) 2009, Willow Garage Inc., all rights reserved.
15 // Copyright (C) 2013, OpenCV Foundation, all rights reserved.
16 // Third party copyrights are property of their respective owners.
18 // Redistribution and use in source and binary forms, with or without modification,
19 // are permitted provided that the following conditions are met:
21 // * Redistribution's of source code must retain the above copyright notice,
22 // this list of conditions and the following disclaimer.
24 // * Redistribution's in binary form must reproduce the above copyright notice,
25 // this list of conditions and the following disclaimer in the documentation
26 // and/or other materials provided with the distribution.
28 // * The name of the copyright holders may not be used to endorse or promote products
29 // derived from this software without specific prior written permission.
31 // This software is provided by the copyright holders and contributors "as is" and
32 // any express or implied warranties, including, but not limited to, the implied
33 // warranties of merchantability and fitness for a particular purpose are disclaimed.
34 // In no event shall the Intel Corporation or contributors be liable for any direct,
35 // indirect, incidental, special, exemplary, or consequential damages
36 // (including, but not limited to, procurement of substitute goods or services;
37 // loss of use, data, or profits; or business interruption) however caused
38 // and on any theory of liability, whether in contract, strict liability,
39 // or tort (including negligence or otherwise) arising in any way out of
40 // the use of this software, even if advised of the possibility of such damage.
44 #ifndef __OPENCV_OBJDETECT_HPP__
45 #define __OPENCV_OBJDETECT_HPP__
47 #include "opencv2/core.hpp"
49 typedef struct CvLatentSvmDetector CvLatentSvmDetector;
50 typedef struct CvHaarClassifierCascade CvHaarClassifierCascade;
55 ///////////////////////////// Object Detection ////////////////////////////
58 * This is a class wrapping up the structure CvLatentSvmDetector and functions working with it.
59 * The class goals are:
60 * 1) provide c++ interface;
61 * 2) make it possible to load and detect more than one class (model) unlike CvLatentSvmDetector.
63 class CV_EXPORTS LatentSvmDetector
66 struct CV_EXPORTS ObjectDetection
69 ObjectDetection( const Rect& rect, float score, int classID = -1 );
76 LatentSvmDetector( const std::vector<String>& filenames, const std::vector<String>& classNames = std::vector<String>() );
77 virtual ~LatentSvmDetector();
80 virtual bool empty() const;
81 bool load( const std::vector<String>& filenames, const std::vector<String>& classNames = std::vector<String>() );
83 virtual void detect( const Mat& image,
84 std::vector<ObjectDetection>& objectDetections,
85 float overlapThreshold = 0.5f,
86 int numThreads = -1 );
88 const std::vector<String>& getClassNames() const;
89 size_t getClassCount() const;
92 std::vector<CvLatentSvmDetector*> detectors;
93 std::vector<String> classNames;
96 // class for grouping object candidates, detected by Cascade Classifier, HOG etc.
97 // instance of the class is to be passed to cv::partition (see cxoperations.hpp)
98 class CV_EXPORTS SimilarRects
101 SimilarRects(double _eps) : eps(_eps) {}
102 inline bool operator()(const Rect& r1, const Rect& r2) const
104 double delta = eps*(std::min(r1.width, r2.width) + std::min(r1.height, r2.height))*0.5;
105 return std::abs(r1.x - r2.x) <= delta &&
106 std::abs(r1.y - r2.y) <= delta &&
107 std::abs(r1.x + r1.width - r2.x - r2.width) <= delta &&
108 std::abs(r1.y + r1.height - r2.y - r2.height) <= delta;
113 CV_EXPORTS void groupRectangles(std::vector<Rect>& rectList, int groupThreshold, double eps = 0.2);
114 CV_EXPORTS_W void groupRectangles(CV_IN_OUT std::vector<Rect>& rectList, CV_OUT std::vector<int>& weights,
115 int groupThreshold, double eps = 0.2);
116 CV_EXPORTS void groupRectangles(std::vector<Rect>& rectList, int groupThreshold,
117 double eps, std::vector<int>* weights, std::vector<double>* levelWeights );
118 CV_EXPORTS void groupRectangles(std::vector<Rect>& rectList, std::vector<int>& rejectLevels,
119 std::vector<double>& levelWeights, int groupThreshold, double eps = 0.2);
120 CV_EXPORTS void groupRectangles_meanshift(std::vector<Rect>& rectList, std::vector<double>& foundWeights,
121 std::vector<double>& foundScales,
122 double detectThreshold = 0.0, Size winDetSize = Size(64, 128));
124 template<> CV_EXPORTS void DefaultDeleter<CvHaarClassifierCascade>::operator ()(CvHaarClassifierCascade* obj) const;
126 enum { CASCADE_DO_CANNY_PRUNING = 1,
127 CASCADE_SCALE_IMAGE = 2,
128 CASCADE_FIND_BIGGEST_OBJECT = 4,
129 CASCADE_DO_ROUGH_SEARCH = 8
132 class CV_EXPORTS_W BaseCascadeClassifier : public Algorithm
135 virtual ~BaseCascadeClassifier();
136 virtual bool empty() const = 0;
137 virtual bool load( const String& filename ) = 0;
138 virtual void detectMultiScale( InputArray image,
139 CV_OUT std::vector<Rect>& objects,
141 int minNeighbors, int flags,
142 Size minSize, Size maxSize ) = 0;
144 virtual void detectMultiScale( InputArray image,
145 CV_OUT std::vector<Rect>& objects,
146 CV_OUT std::vector<int>& numDetections,
148 int minNeighbors, int flags,
149 Size minSize, Size maxSize ) = 0;
151 virtual void detectMultiScale( InputArray image,
152 CV_OUT std::vector<Rect>& objects,
153 CV_OUT std::vector<int>& rejectLevels,
154 CV_OUT std::vector<double>& levelWeights,
156 int minNeighbors, int flags,
157 Size minSize, Size maxSize,
158 bool outputRejectLevels ) = 0;
160 virtual bool isOldFormatCascade() const = 0;
161 virtual Size getOriginalWindowSize() const = 0;
162 virtual int getFeatureType() const = 0;
163 virtual void* getOldCascade() = 0;
165 class CV_EXPORTS MaskGenerator
168 virtual ~MaskGenerator() {}
169 virtual Mat generateMask(const Mat& src)=0;
170 virtual void initializeMask(const Mat& /*src*/) { }
172 virtual void setMaskGenerator(const Ptr<MaskGenerator>& maskGenerator) = 0;
173 virtual Ptr<MaskGenerator> getMaskGenerator() = 0;
176 class CV_EXPORTS_W CascadeClassifier
179 CV_WRAP CascadeClassifier();
180 CV_WRAP CascadeClassifier(const String& filename);
181 ~CascadeClassifier();
182 CV_WRAP bool empty() const;
183 CV_WRAP bool load( const String& filename );
184 CV_WRAP bool read( const FileNode& node );
185 CV_WRAP void detectMultiScale( InputArray image,
186 CV_OUT std::vector<Rect>& objects,
187 double scaleFactor = 1.1,
188 int minNeighbors = 3, int flags = 0,
189 Size minSize = Size(),
190 Size maxSize = Size() );
192 CV_WRAP_AS(detectMultiScale2) void detectMultiScale( InputArray image,
193 CV_OUT std::vector<Rect>& objects,
194 CV_OUT std::vector<int>& numDetections,
195 double scaleFactor=1.1,
196 int minNeighbors=3, int flags=0,
198 Size maxSize=Size() );
200 CV_WRAP_AS(detectMultiScale3) void detectMultiScale( InputArray image,
201 CV_OUT std::vector<Rect>& objects,
202 CV_OUT std::vector<int>& rejectLevels,
203 CV_OUT std::vector<double>& levelWeights,
204 double scaleFactor = 1.1,
205 int minNeighbors = 3, int flags = 0,
206 Size minSize = Size(),
207 Size maxSize = Size(),
208 bool outputRejectLevels = false );
210 CV_WRAP bool isOldFormatCascade() const;
211 CV_WRAP Size getOriginalWindowSize() const;
212 CV_WRAP int getFeatureType() const;
213 void* getOldCascade();
215 CV_WRAP static bool convert(const String& oldcascade, const String& newcascade);
217 void setMaskGenerator(const Ptr<BaseCascadeClassifier::MaskGenerator>& maskGenerator);
218 Ptr<BaseCascadeClassifier::MaskGenerator> getMaskGenerator();
220 Ptr<BaseCascadeClassifier> cc;
223 CV_EXPORTS Ptr<BaseCascadeClassifier::MaskGenerator> createFaceDetectionMaskGenerator();
225 //////////////// HOG (Histogram-of-Oriented-Gradients) Descriptor and Object Detector //////////////
227 // struct for detection region of interest (ROI)
230 // scale(size) of the bounding box
232 // set of requrested locations to be evaluated
233 std::vector<cv::Point> locations;
234 // vector that will contain confidence values for each location
235 std::vector<double> confidences;
238 struct CV_EXPORTS_W HOGDescriptor
243 enum { DEFAULT_NLEVELS = 64
246 CV_WRAP HOGDescriptor() : winSize(64,128), blockSize(16,16), blockStride(8,8),
247 cellSize(8,8), nbins(9), derivAperture(1), winSigma(-1),
248 histogramNormType(HOGDescriptor::L2Hys), L2HysThreshold(0.2), gammaCorrection(true),
249 free_coef(-1.f), nlevels(HOGDescriptor::DEFAULT_NLEVELS)
252 CV_WRAP HOGDescriptor(Size _winSize, Size _blockSize, Size _blockStride,
253 Size _cellSize, int _nbins, int _derivAperture=1, double _winSigma=-1,
254 int _histogramNormType=HOGDescriptor::L2Hys,
255 double _L2HysThreshold=0.2, bool _gammaCorrection=false,
256 int _nlevels=HOGDescriptor::DEFAULT_NLEVELS)
257 : winSize(_winSize), blockSize(_blockSize), blockStride(_blockStride), cellSize(_cellSize),
258 nbins(_nbins), derivAperture(_derivAperture), winSigma(_winSigma),
259 histogramNormType(_histogramNormType), L2HysThreshold(_L2HysThreshold),
260 gammaCorrection(_gammaCorrection), free_coef(-1.f), nlevels(_nlevels)
263 CV_WRAP HOGDescriptor(const String& filename)
268 HOGDescriptor(const HOGDescriptor& d)
273 virtual ~HOGDescriptor() {}
275 CV_WRAP size_t getDescriptorSize() const;
276 CV_WRAP bool checkDetectorSize() const;
277 CV_WRAP double getWinSigma() const;
279 CV_WRAP virtual void setSVMDetector(InputArray _svmdetector);
281 virtual bool read(FileNode& fn);
282 virtual void write(FileStorage& fs, const String& objname) const;
284 CV_WRAP virtual bool load(const String& filename, const String& objname = String());
285 CV_WRAP virtual void save(const String& filename, const String& objname = String()) const;
286 virtual void copyTo(HOGDescriptor& c) const;
288 CV_WRAP virtual void compute(InputArray img,
289 CV_OUT std::vector<float>& descriptors,
290 Size winStride = Size(), Size padding = Size(),
291 const std::vector<Point>& locations = std::vector<Point>()) const;
293 //with found weights output
294 CV_WRAP virtual void detect(const Mat& img, CV_OUT std::vector<Point>& foundLocations,
295 CV_OUT std::vector<double>& weights,
296 double hitThreshold = 0, Size winStride = Size(),
297 Size padding = Size(),
298 const std::vector<Point>& searchLocations = std::vector<Point>()) const;
299 //without found weights output
300 virtual void detect(const Mat& img, CV_OUT std::vector<Point>& foundLocations,
301 double hitThreshold = 0, Size winStride = Size(),
302 Size padding = Size(),
303 const std::vector<Point>& searchLocations=std::vector<Point>()) const;
305 //with result weights output
306 CV_WRAP virtual void detectMultiScale(InputArray img, CV_OUT std::vector<Rect>& foundLocations,
307 CV_OUT std::vector<double>& foundWeights, double hitThreshold = 0,
308 Size winStride = Size(), Size padding = Size(), double scale = 1.05,
309 double finalThreshold = 2.0,bool useMeanshiftGrouping = false) const;
310 //without found weights output
311 virtual void detectMultiScale(InputArray img, CV_OUT std::vector<Rect>& foundLocations,
312 double hitThreshold = 0, Size winStride = Size(),
313 Size padding = Size(), double scale = 1.05,
314 double finalThreshold = 2.0, bool useMeanshiftGrouping = false) const;
316 CV_WRAP virtual void computeGradient(const Mat& img, CV_OUT Mat& grad, CV_OUT Mat& angleOfs,
317 Size paddingTL = Size(), Size paddingBR = Size()) const;
319 CV_WRAP static std::vector<float> getDefaultPeopleDetector();
320 CV_WRAP static std::vector<float> getDaimlerPeopleDetector();
322 CV_PROP Size winSize;
323 CV_PROP Size blockSize;
324 CV_PROP Size blockStride;
325 CV_PROP Size cellSize;
327 CV_PROP int derivAperture;
328 CV_PROP double winSigma;
329 CV_PROP int histogramNormType;
330 CV_PROP double L2HysThreshold;
331 CV_PROP bool gammaCorrection;
332 CV_PROP std::vector<float> svmDetector;
338 // evaluate specified ROI and return confidence value for each location
339 virtual void detectROI(const cv::Mat& img, const std::vector<cv::Point> &locations,
340 CV_OUT std::vector<cv::Point>& foundLocations, CV_OUT std::vector<double>& confidences,
341 double hitThreshold = 0, cv::Size winStride = Size(),
342 cv::Size padding = Size()) const;
344 // evaluate specified ROI and return confidence value for each location in multiple scales
345 virtual void detectMultiScaleROI(const cv::Mat& img,
346 CV_OUT std::vector<cv::Rect>& foundLocations,
347 std::vector<DetectionROI>& locations,
348 double hitThreshold = 0,
349 int groupThreshold = 0) const;
351 // read/parse Dalal's alt model file
352 void readALTModel(String modelfile);
353 void groupRectangles(std::vector<cv::Rect>& rectList, std::vector<double>& weights, int groupThreshold, double eps) const;
357 CV_EXPORTS_W void findDataMatrix(InputArray image,
358 CV_OUT std::vector<String>& codes,
359 OutputArray corners = noArray(),
360 OutputArrayOfArrays dmtx = noArray());
362 CV_EXPORTS_W void drawDataMatrixCodes(InputOutputArray image,
363 const std::vector<String>& codes,
367 #include "opencv2/objdetect/linemod.hpp"
368 #include "opencv2/objdetect/erfilter.hpp"