export to python/java
This commit is contained in:
parent
665bf430d5
commit
bd9ca48fab
@ -39,6 +39,7 @@ Implementation of soft (stageless) cascaded detector. ::
|
||||
cv::AlgorithmInfo* info() const;
|
||||
virtual bool load(const FileNode& fn);
|
||||
virtual void detect(InputArray image, InputArray rois, std::vector<Detection>& objects) const;
|
||||
virtual void detect(InputArray image, InputArray rois, OutputArray rects, OutputArray confs) const;
|
||||
};
|
||||
|
||||
|
||||
@ -80,10 +81,16 @@ SCascade::detect
|
||||
--------------------------
|
||||
Apply cascade to an input frame and return the vector of Decection objcts.
|
||||
|
||||
.. ocv:function:: bool SCascade::detect(InputArray image, InputArray rois, std::vector<Detection>& objects) const
|
||||
.. ocv:function:: void SCascade::detect(InputArray image, InputArray rois, std::vector<Detection>& objects) const
|
||||
|
||||
.. ocv:function:: void SCascade::detect(InputArray image, InputArray rois, OutputArray rects, OutputArray confs) const
|
||||
|
||||
:param image: a frame on which detector will be applied.
|
||||
|
||||
:param rois: a vector of regions of interest. Only the objects that fall into one of the regions will be returned.
|
||||
|
||||
:param objects: an output array of Detections.
|
||||
:param objects: an output array of Detections.
|
||||
|
||||
:param rects: an output array of bounding rectangles for detected objects.
|
||||
|
||||
:param confs: an output array of confidence for detected objects. i-th bounding rectangle corresponds i-th configence.
|
@ -490,7 +490,7 @@ protected:
|
||||
|
||||
|
||||
// Implementation of soft (stageless) cascaded detector.
|
||||
class CV_EXPORTS SCascade : public Algorithm
|
||||
class CV_EXPORTS_W SCascade : public Algorithm
|
||||
{
|
||||
public:
|
||||
|
||||
@ -539,24 +539,27 @@ public:
|
||||
// Param minScale is a maximum scale relative to the original size of the image on which cascade will be applyed.
|
||||
// Param scales is a number of scales from minScale to maxScale.
|
||||
// Param rejfactor is used for NMS.
|
||||
SCascade(const double minScale = 0.4, const double maxScale = 5., const int scales = 55, const int rejfactor = 1);
|
||||
CV_WRAP SCascade(const double minScale = 0.4, const double maxScale = 5., const int scales = 55, const int rejfactor = 1);
|
||||
|
||||
virtual ~SCascade();
|
||||
CV_WRAP virtual ~SCascade();
|
||||
|
||||
cv::AlgorithmInfo* info() const;
|
||||
|
||||
// Load cascade from FileNode.
|
||||
// Param fn is a root node for cascade. Should be <cascade>.
|
||||
virtual bool load(const FileNode& fn);
|
||||
CV_WRAP virtual bool load(const FileNode& fn);
|
||||
|
||||
// Load cascade config.
|
||||
virtual void read(const FileNode& fn);
|
||||
CV_WRAP virtual void read(const FileNode& fn);
|
||||
|
||||
// Return the vector of Decection objcts.
|
||||
// Param image is a frame on which detector will be applied.
|
||||
// Param rois is a vector of regions of interest. Only the objects that fall into one of the regions will be returned.
|
||||
// Param objects is an output array of Detections
|
||||
virtual void detect(InputArray image, InputArray rois, std::vector<Detection>& objects) const;
|
||||
// Param rects is an output array of bounding rectangles for detected objects.
|
||||
// Param confs is an output array of confidence for detected objects. i-th bounding rectangle corresponds i-th configence.
|
||||
CV_WRAP virtual void detect(InputArray image, InputArray rois, OutputArray rects, OutputArray confs) const;
|
||||
|
||||
private:
|
||||
void detectNoRoi(const Mat& image, std::vector<Detection>& objects) const;
|
||||
|
@ -500,4 +500,27 @@ void cv::SCascade::detect(cv::InputArray _image, cv::InputArray _rois, std::vect
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void cv::SCascade::detect(InputArray _image, InputArray _rois, OutputArray _rects, OutputArray _confs) const
|
||||
{
|
||||
std::vector<Detection> objects;
|
||||
detect( _image, _rois, objects);
|
||||
|
||||
_rects.create(1, objects.size(), CV_32SC4);
|
||||
cv::Mat_<cv::Rect> rects = (cv::Mat_<cv::Rect>)_rects.getMat();
|
||||
cv::Rect* rectPtr = rects.ptr<cv::Rect>(0);
|
||||
|
||||
_confs.create(1, objects.size(), CV_32F);
|
||||
cv::Mat confs = _confs.getMat();
|
||||
float* confPtr = rects.ptr<float>(0);
|
||||
|
||||
typedef std::vector<Detection>::const_iterator IDet;
|
||||
|
||||
int i = 0;
|
||||
for (IDet it = objects.begin(); it != objects.end(); ++it, ++i)
|
||||
{
|
||||
rectPtr[i] = (*it).bb;
|
||||
confPtr[i] = (*it).confidence;
|
||||
}
|
||||
}
|
@ -66,34 +66,26 @@ TEST(SCascade, detect)
|
||||
std::vector<Detection> objects;
|
||||
|
||||
cascade.detect(colored, cv::noArray(), objects);
|
||||
|
||||
// cv::Mat out = colored.clone();
|
||||
// int level = 0, total = 0;
|
||||
// int levelWidth = objects[0].bb.width;
|
||||
|
||||
// for(int i = 0 ; i < (int)objects.size(); ++i)
|
||||
// {
|
||||
// if (objects[i].bb.width != levelWidth)
|
||||
// {
|
||||
// std::cout << "Level: " << level << " total " << total << std::endl;
|
||||
// cv::imshow("out", out);
|
||||
// cv::waitKey(0);
|
||||
// out = colored.clone();
|
||||
// levelWidth = objects[i].bb.width;
|
||||
// total = 0;
|
||||
// level++;
|
||||
// }
|
||||
// cv::rectangle(out, objects[i].bb, cv::Scalar(255, 0, 0, 255), 1);
|
||||
// std::cout << "detection: " << objects[i].bb.x
|
||||
// << " " << objects[i].bb.y
|
||||
// << " " << objects[i].bb.width
|
||||
// << " " << objects[i].bb.height << std::endl;
|
||||
// total++;
|
||||
// }
|
||||
// std::cout << "detected: " << (int)objects.size() << std::endl;
|
||||
ASSERT_EQ((int)objects.size(), 3498);
|
||||
}
|
||||
|
||||
TEST(SCascade, detectSeparate)
|
||||
{
|
||||
typedef cv::SCascade::Detection Detection;
|
||||
std::string xml = cvtest::TS::ptr()->get_data_path() + "cascadeandhog/sc_cvpr_2012_to_opencv.xml";
|
||||
cv::SCascade cascade;
|
||||
cv::FileStorage fs(xml, cv::FileStorage::READ);
|
||||
ASSERT_TRUE(cascade.load(fs.getFirstTopLevelNode()));
|
||||
|
||||
cv::Mat colored = cv::imread(cvtest::TS::ptr()->get_data_path() + "cascadeandhog/bahnhof/image_00000000_0.png");
|
||||
ASSERT_FALSE(colored.empty());
|
||||
|
||||
cv::Mat rects, confs;
|
||||
|
||||
cascade.detect(colored, cv::noArray(), rects, confs);
|
||||
ASSERT_EQ(confs.cols, 3498);
|
||||
}
|
||||
|
||||
TEST(SCascade, detectRoi)
|
||||
{
|
||||
typedef cv::SCascade::Detection Detection;
|
||||
@ -110,31 +102,6 @@ TEST(SCascade, detectRoi)
|
||||
rois.push_back(cv::Rect(0, 0, 640, 480));
|
||||
|
||||
cascade.detect(colored, rois, objects);
|
||||
|
||||
// cv::Mat out = colored.clone();
|
||||
// int level = 0, total = 0;
|
||||
// int levelWidth = objects[0].bb.width;
|
||||
|
||||
// for(int i = 0 ; i < (int)objects.size(); ++i)
|
||||
// {
|
||||
// if (objects[i].bb.width != levelWidth)
|
||||
// {
|
||||
// std::cout << "Level: " << level << " total " << total << std::endl;
|
||||
// cv::imshow("out", out);
|
||||
// cv::waitKey(0);
|
||||
// out = colored.clone();
|
||||
// levelWidth = objects[i].bb.width;
|
||||
// total = 0;
|
||||
// level++;
|
||||
// }
|
||||
// cv::rectangle(out, objects[i].bb, cv::Scalar(255, 0, 0, 255), 1);
|
||||
// std::cout << "detection: " << objects[i].bb.x
|
||||
// << " " << objects[i].bb.y
|
||||
// << " " << objects[i].bb.width
|
||||
// << " " << objects[i].bb.height << std::endl;
|
||||
// total++;
|
||||
// }
|
||||
// std::cout << "detected: " << (int)objects.size() << std::endl;
|
||||
ASSERT_EQ((int)objects.size(), 3498);
|
||||
}
|
||||
|
||||
|
Loading…
x
Reference in New Issue
Block a user