Use of Hough circles in OpenCV

OpenCV provides and inbuilt function to detect circles in your image. The Hough transform is a technique which can be used to isolate features of a particular shape within an image. The implementation of the circle  transform in OpenCV uses a tricky method called the Hough gradient method ( Link to Google Books ).

cvHoughCircles( ) is the function provided in OpenCV.

CvSeq* cvHoughCircles( 
        CvArr* image,         
        void*  circle_storage, 
        int    method, 
        double dp, 
        double min_dist, 
        double param1 = 100, 
        double param2 = 300, 
        int    min_radius = 0, 
        int    max_radius = 0 

Following is a example code snippet.

//  Allocating memory dynamically to store the circle regions detected
CvMemStorage* storage_var = cvCreateMemStorage(0);

//  A sequence of circles will get stored in the results variable
CvSeq* results = cvHoughCircles(img, storage_var , CV_HOUGH_GRADIENT , 2 , img->height/3 );

//  Using a for loop to access individual circles; draw a circle in the input image
for( int i = 0; i < results->total; i++ )
      float* p = (float*) cvGetSeqElem( results, i );
      CvPoint pt = cvPoint( cvRound( p[0] ), cvRound( p[1] ));
      cvCircle(img,pt,cvRound( p[2] ),cvScalar(0,0,255),1.8);

Some things to note before using cvHoughTransform( )
  • The image used should be in grey-scale.
  • Reduce noise in the image as much as possible. You can use cvSmooth( ) function with CV_GRADIENT or CV_MEDIAN types to achieve this. Also many other filters could be used depending on the image. 
cvHoughCircles( ) is a very inefficient method! I had tried using it to detect the pupil in the eyes. It was a very lame try because pupil detection actually turned out to be a very complex procedure! I would like to share some of the experiments i tried by implementing hough transform.

Initially i converted image to grey scale and applied gaussian smooth. Implemeted cvHoughCircles( ) on this image. You can see that the circle detection has been all over the place. Even the pupil which i wanted to detect is not accurate! So basically i had to filter more and reduce features in the image. 

So next i just increased the intensity levels of the pixels. I can say that's one method which worked for me in this case. The number of circles detected was reduced and also a more accurate pupil boundary was detected. Similarly you can try various other filters.


Андрей said...
This comment has been removed by the author.
Adithya said...

Ya right..

ratna paul said...

Can you give me some details regarding the filters we can use ??

Exorcismus said...

try median filter , or gaussian

Harry said...

Hi, Nice post! Would you please consider adding a link to my website on your page. Please email me back.



randy said...

Hi, Nice post thanks for sharing. Would you please consider adding a link to my website on your page. Please email me back.



Suraj C K said...

How will i access the source code of the function HoughCircles?

Post a Comment