Abstract:In order to solve the problem of automatic online detection of the size,position,direction and defects of various micro components,an intelligent online detection system based on machine vision and deep learning was proposed.Through the established visual inspection system,the images of micro components were collected;image preprocessing,binarization,filtering,edge contour feature extraction and template matching were completed;the size,position and direction of a variety of micro components were online detected.Aiming at the surface defects of micro components,a method based on deep learning was proposed.The experimental results show that the system can be compatible with a variety of products online detection,the detection efficiency is about 0.344 s/piece,the accuracy rate of size,direction and position detection is 100%,and the accuracy rate of defect identification is about 95.56%.