Hog & SVM

The hog and SVM methods are already very classic and widely used in a variety of industries and integrated into hardware. It should be said that as an important method of motion detection, it is still widely used. (Opencv built-in classifier) int main () {mat src = imread ("1.png"); hogdescriptor hog; // hog Feature Detector hog. setsvmdetector (hogdescriptor: getdefaultpeopledetector (); // sets SVM classifier as the default parameter vector <rect> found, found_filtered; // an array of rectangular boxes hog. detectmultiscale (SRC, found, 0, size (8, 8), size (32, 32), 1.05, 2); // perform multi-scale cout detection on the image <"Number of rectangles: "<found. size () <Endl; // locate all rectangle frames without nesting R, and put them in found_filtered. If nested, put the largest rectangle outside into found_filtered. For (INT I = 0; I <found. size (); I ++) {rect r = found [I]; Int J = 0; For (; j <found. size (); j ++) if (J! = I & (R & found [J]) = r) break; If (j = found. size () found_filtered.push_back (r);} cout <"Number of rectangles after filtering:" <found_filtered.size () <Endl; // draws a rectangle, because the rectangular frame detected by hog is slightly larger than the actual body frame, you need to make some adjustments here (INT I = 0; I <found_filtered.size (); I ++) {rect r = found_filtered [I]; R. X + = cvround (R. width * 0.1); R. width = cvround (R. width * 0.8); R. Y + = cvround (R. height * 0.07); R. height = cvround (R. height * 0.8); rectangle (SRC, R. tl (), R.br (), scalar (0,255, 0), 3);} imwrite ("imgprocessed.jpg", Src); namedwindow ("src", 0 ); imshow ("src", Src); waitkey (); // Note: waitkey must be added after imshow; otherwise, the image system ("pause") cannot be displayed ");} what is embodied here is mainly the usage of the classifier that has been successfully used. In this hog & SVM, it is most important to find the correct hog feature and identify the correct SVM. The effect is very general.