Circle and circular arc detection algorithm research based on Freeman chain code

To detect the information of circle and circular arc, a method based on Freeman chain code combined with segment detection algorithm was proposed. According to the properties of circle Freeman chain code, full circle will be detected at first. Sequently, the rest of edge information belonging to the full circle will be dealt with. Finally, the circular arc detection will be finished by fitting and clustering the edge information by segment under proper condition setting in advance. The result indicated that the proposed method is suitable for detecting the circle with large missing arc and local distortion. Comparing to other algorithms, it is capable of reducing missing detection and providing reliable result with speediness and good robustness.