The Quality Of The Precious Metals Complex With Welding Machine Vision-based Online Detection Technology

Precious metal clad is widely used in contactors, relays, thermostats, timers, plate switches, circuit breakers and other electrical equipments. The purpose of this dissertation is to research and realize an on-line detection system which could detect the welding precision of precious metal clad based on the requirements of the automated welding production line for precious metal clad from Zhongyi Technology Co., Ltd. Dongguan, Guangdong.Based on the theory of machine vision and the reality of the automatic welding production line, and considered the integrity, scalability, openness and reliability of a system, this dissertation proposes a dynamic and flexible on-line detection system whose original target is semi-finished products.This dissertation analyzes the production process and common defects of the precious metal clad. The hardware platform which is built up by reasonable equipments has been proposed in this dissertation to meet the needs of the actual demand. From the theory of machine vision, this dissertation analyzes and realizes the algorithms of image processing which contains image histogram modification, image gray-scale transformation, image filtering, image segmentation, edge detection and image recognition. Further more, the detecting algorithms of three kinds of welding vices are also given in this dissertation. Based on the detecting algorithms, the detecting process of welding vices is designed. This dissertation also realizes the software whose functions is defined by detecting process for the on-line detection system with the help of C# and Halcon. Compared with the results of three different kinds of precious metal clad which is checked by human and machine, the welding vices on-line detection system is proved to be very effective.However, the dynamic and flexible on-line detection system whose original target is semi-finished products could greatly improve the detection efficiency, reduce the labor intensity, find the welding vices before the followed production process and increase the flexibility of the automated welding production line. What's more, the system proposed in this dissertation will be the develop trend of the on-line detection system. The on-line detection system of precious metal clad which based on machine vision is going to be used in production, there is no doubt that it will creat great benefits both in economy and society.