Machine Vision-Based Detection Method for Surface Micro-Defect of the Machined Parts

In order to improve the precision of micro-defect detection results and improve the appearance quality of machined parts, the machine vision technology is introduced in this paper. Taking a mechanical manufacturing unit as an example, a new technical method for micro-defects on the surface of parts is designed. According to the application requirements of machine vision technology, a scanning device integrating industrial camera, acquisition device and illumination light source is built to collect the surface image of parts. The collected original image is processed by mean filtering to remove the interference of relevant factors and noise in the image on identification of the defect areas. The method of threshold segmentation is adopted to extract and divide the micro-defect areas on the surface of machined parts. The method of extracting image edge operator is used to calculate the pixel correlation between the original image of the part surface and the image to be detected, and design of the detection method is completed by matching the gray nature points of the part surface micro-defects. The comparative experimental results show that this method can not only accurately detect the surface micro-defects of machined parts, but also detect the types of defects.