A Detection and Identification Method Based on Machine Vision for Bearing Surface Defects

In view of the disadvantages of manual testing of bearing surface defects in the bearing production process, an automatic detection and identification method of bearing surface defects based on machine vision is proposed. Firstly, the source image is pre-processed by gamma correction algorithm, and the Canny algorithm is improved by adaptive selection of the Canny algorithm based on iterative threshold segmentation method and Ostu algorithm to improve the integrity and precision of the segmentation of bearing surface defects. The experimental results show that the method can be accurately detected of bearing surface defects, and the defect recognition rate has reached 93.33%.