opencv 霍夫变换检测圆cvHoughCircles

 CVAPI(CvSeq*) cvHoughCircles( CvArr* image, void* circle_storage,
                               int method, double dp, double min_dist,
                               double param1 CV_DEFAULT(100),
                               double param2 CV_DEFAULT(100),
                               int min_radius CV_DEFAULT(0),
                               int max_radius CV_DEFAULT(0));

image  The 8-bit, single-channel, grayscale input image 


circles  The output vector of found circles.  Each vector is encoded as 3-element floating-point 
      vector    


method    Currently, the only implemented method is CV HOUGH GRADIENT, which is basically 21HT, 
      described in [25]. 


dp  The inverse ratio of the accumulator resolution to the image resolution.  For example, if dp=1, 
      the accumulator will have the same resolution as the input image, if dp=2 - accumulator will 
      have half as big width and height, etc 


minDist    Minimum distance between the centers of the detected circles.  If the parameter is too 
      small, multiple neighbor circles may be falsely detected in addition to a true one.  If it is too 
      large, some circles may be missed 

param1   The first method-specific parameter. in the case of CV HOUGH GRADIENT it is the higher 
      threshold of the two passed to cv::Canny edge detector (the lower one will be twice smaller) 


param2   The  second  method-specific  parameter.    in  the  case  of  CV HOUGH GRADIENT  it  is  the 
      accumulator threshold at the center detection stage. The smaller it is, the more false circles 
      may be detected.  Circles, corresponding to the larger accumulator values, will be returned 
      first 


minRadius     Minimum circle radius 


maxRadius     Maximum circle radius 

转自http://blog.csdn.net/fdl19881/article/details/6714530