Circle Detection Using a Gabor Annulus

We present a novel circle detection technique based on the desirable properties of Gabor wavelet filters. We describe the technique and its Gabor wavelet origins, and perform multiple experiments using both synthetic images and real microscopic images of red blood cells. Our experiments show the proposed technique’s ability to detect single and overlapping circles, with improved results over traditional techniques including both the Circular Hough Transform and its modifications which include edge orientation. The improved results are demonstrated under high levels of noise and for partially defined overlapping circles.