The Research Of Small Parts Quality Detection Method And System Based On Machine Vision

With the development of science and technology, electronic products tend to miniaturization and higher requirements have been put forward for the quality inspection accordingly. The traditional methods to inspect the quality of some small parts are mostly to artificially inspect them with measurement tools(such as micrometer, microscope, etc.). However, the artificial detection methods have limitations of low efficiency, high rate of missed detection, high false positive rate and less detection precision etc. These methods have been gradually taken place by non-contact detection methods, such as laser detection, radiographic inspection and machine visual inspection and so on. On the basis of machine vision, the non-contact detection methods which have the advantages of high speed, high precision, high efficiency and less effort are widely used in industry monitor, product quality test and PCB defect detection and satellite reconnaissance.This paper analyses the characteristics of small parts, including geometry features, surface quality and testability and forward carries out the detection methods of small parts based on machine vision. This paper introduces the general idea, technical principle, testing process, the pixel size equivalent calibration method, light intensity of illumination compensation method, macro automatic focus detection and scanning method and so on. And based these calculation, the detection algorithm of the sizes and surface defect based on monocular CCD are presented. By non-contact means to achieve geometry automatic measure and surface defects of small parts for PCB micro-drill, plastic parts and glass panel. This paper designs the quality inspection system to inspect small parts based on machine vision, which is used to quality testing of PCB micro-drill, plastic parts and glass panel.The test result of the algorithm simulation and system operation indicates that the relevant quality inspection algorithm and key technology to inspect small parts based on machine vision in this paper are correct and feasible. The above methods can achieve online automatic detection including the overall length and edge diameter of PCB micro-drill, scratch detect inspection of plastic parts of automobiles and glass panel. And they have high detection precision and efficiency. It has an important practical application value for improving the production efficiency and quality of PCB micro-drill, plastic parts and glass panel.