Research And Implementation Of Printing Defect Detection System Based On Machine Vision

Terminal blocks which are used to connect wires in the circuit are mass produced to meet the needs of circuit connection. However, in large-scale production process, quality inspection of the characters is an indispensable link. Through the research, we find that the quality of printed characters on terminal blocks is basically detected by artificial vision in SMES. The cost of that detection method is high and the quality is not guaranteed, which restricts high-efficient production in the enterprise. Therefore, the automatic detection system for the quality inspection of the characters is to be built to solve the urgent need of the enterprise, which can check the quality of the characters efficiently, fast and automatically.Therefore, this paper focuses on the research and designation of the printing defect detection system which is based on the machine vision. The main work items are as follows.(1) The on-site research of the production and processing about terminal blocks is performed, and defect types of printing characters are analyzed;(2) The experimental platform is set up to real-timely capture the images of the terminal blocks. According to the characteristics of the terminal blocks, the lighting system is designed, which meets the detection condition, besides, the camera, lens and industrial computer are selected for the detection.(3) The experiments using the captured images of the terminal blocks are carried on. Image smoothing, image correction, feature extraction is performed to prepare for defect analysis.(4) Based on the pre-processed images, the characteristics of the qualified images and printing defect images are analyzed, and the defects detection algorithm of the terminal blocks is designed, besides, simulation experiment is performed successfully under IMAQ Vision.(5) By using C #.Net and IMQA Vision image development kits, terminals print defect detection system is developed. Test aimed at the hardware, algorithms and software of the detection system is carried on. Test results show that: the system achieves the expected purpose of the research and it can be applied to detect printing terminals.