Detection of Circular Arcs within Tolerant Error
Sung-Pil Lyu · Jeongbo gwahaghoe nonmunji. so'peuteuweeo mich eung'yong · 2005
Arcs are usually treated as significant features in the field of pattern recognition. This paper presents a method to detect arcs from digital planar curves and estimate their arc centers and radii by using geometric analysis. The deviation of distance between the original curve and the detected arc by the proposed method does not exceed a tolerant error. The experimental results show that the proposed method is available for the detection of arcs iron not only smooth but also heavily noisy curves.