A Fast Recognition Method of Workpiece Mark Points Based on Machine Vision
Yiquan Hao, Jinyu Guo · 2025
In response to the problems of lack of accuracy in the recognition process, this paper proposes a fast recognition method for workpiece Mark points based on machine vision. First, an industrial camera is used to capture the original fixture image, and pre - processing operations are employed to reduce the influence of image noise and jagged edges. Then, Canny edge extraction is performed on the fixture, followed by the acquisition of the sub - pixel edge point set. Finally, Radon transform circle detection and least - squares circular fitting are used to locate the Mark points. Experiments show that this method can quickly and accurately obtain the corresponding Mark points, with a positioning accuracy reaching the sub - pixel level and high stability.