Find Corner by SubPix

The location of the corner of digital image processing.
void cvFindCornerSubPix (const CvArr * image, CvPoint2D32f * corners,.
int count, CvSize win, CvSize zero_zone,.
CvTermCriteria criteria);

image = image size as the original. floating-point 32-bit 1 channel.
corners = the value of the position angle of the input image. To locate the point of the output image.
count = number of runs in any of the points after round.
win = half the size of the desired size windows for example, if win = 5, so the size of the windows to find the 5 +5 +1 = 11 (to be positive because it includes a point to begin with) will be that size of the win to find = 11 x 11 itself.
zero_zone = the value to 0 to not want to be thinking.
criteria = highest value after the loop according to the count.

Harris edge detector

Find an edge image by using Harris edge detector.

void cvCornerHarris (const CvArr * image, CvArr * harris_responce,.
int block_size, int aperture_size = 3, double k = 0.04);

image = image size as the original. floating-point 32-bit 1 channel.
harris_responce = picture results after finding marginal Harris picture should be sized originals.
block_size = same function cvCornerEigenValsAndVecs.
aperture_size = see Sobel.
k = Harris detector free parameter which is equal to 0.04 default.

//CornerHarris
IplImage *pHarris=cvCreateImage(cvGetSize(pCornerEigen),IPL_DEPTH_32F,1); cvCornerHarris( pCornerEigen, pHarris, 3 , 7, 0.04 );
cvNamedWindow("Harris",CV_WINDOW_AUTOSIZE);
cvShowImage("Harris", pHarris);