A Robust high-precision circular target detection method based on hough transform

The Circular Hough Transform(CHT) is the most common method in image circle detection in machine vision which is widely used in industrial tasks. However, there are irresolvable factors in theory for precisely target localization which can´t lead to locate circular targets with high-precision. In this paper, the robustness of the algorithm is enhanced with the contours tree analysis and fitting of ellipses that improves the location accuracy. The robustness and location accuracy are extremely improved and that are tested in our actual project of circular object detection. So the proposed improvement can be widely used in the practical applications.