作者: Ming Chen , Feng Zhang , Zhenhong Du , Renyi Liu
DOI: 10.1007/S10043-013-0082-6
关键词: Histogram 、 Image processing 、 Randomized Hough transform 、 Scan line 、 Midpoint circle algorithm 、 Midpoint 、 Artificial intelligence 、 Computer vision 、 Real image 、 Radius 、 Computer science
摘要: Circle detection is significant in image processing and pattern recognition. We present a new algorithm for detecting circles, which based on the global geometric symmetry of circles. First, horizontal vertical midpoint histograms edge are obtained by using scan lines. Then, we apply peak-finding to look center circle. The normalized radius histogram finally used verify existence circle extract its radius. Synthetic images with different levels pepper noise real containing several circles have been taken test performance. Experimental results demonstrate that proposed has advantage computational efficiency as compared randomized Hough transform some other algorithms.