Machine Vision Based Automatic Micro-parts Detection System

Xianchuan Yu, Guanyin Gao, Wu He, Jin-dong Xu

Abstract: Traditional micro-parts defect detection methods are based on human observation to find parts defects, which is inefficient and inaccurate. In order to solve these problems, we designed an automatic parts defect detection system based on machine vision, which include parts transfer module, image capture module, image recognition module and parts selection module. Besides, we put forward a method which is suitable for micro-parts. Experiments show that the method works well and can achieve high accuracy as well as high efficiency, which helps reduce the cost of labor, improve efficiency, production quality and automation.