A Design Of AOI About The Defect Detection Based On The Surface Of PCB Bare Board

Printed circuit board(PCB) is widely used in all aspects of the electronic industry, the detection its quality of has been a concern. With the development of PCB products towards high precision, complexity and super thin type, traditional PCB quality inspection technology is unable to meet the needs of development. How to realize the defect detection of PCB surface has become a hot topic of discussion in a quick and efficient way.Automatic optical inspection(AOI), as a new visual inspection method, has been more and more widely applied to the PCB surface defect detection. AOI technology is a combination of electronics, digital image processing, photoelectric detection and computer technology, machine vision technology, in the field of precision testing technology is very developed. Therefore, it is of great practical value to study the application of AOI technology in the PCB bare surface defect detection and to put forward suggestions for improvement.In view of the AOI technology in the PCB bare board surface defect detection research, has done the following work. First, a complete AOI detection system is designed for the surface defect of PCB bare plate, including image acquisition, image preprocessing, image segmentation, defect detection and recognition, etc.Second, in view of the key technology of camera calibration in the vision inspection system, this paper expounds the theory of camera calibration, and the central idea of the existing calibration algorithm. The applicable occasions together with the advantages and disadvantages of each algorithm are given, according to the operation complexity and the precision of calibration.Third, through the algorithm analysis and research of the existing image segmentation, combined with the characteristic of PCB naked image with high contrast and high resolution characteristics, a histogram based double threshold image segmentation technique is presented.Last, for the demand of PCB bare board surface defect detected with high accuracy, to get the bare PCB images by high-performance light source by a plurality of linear coaxial line scan camera combined with the characteristics of high resolution and high contrast, a new PCB defect detection algorithm based on contour contrast is proposed. The common defects of PCB, such as short circuit, open circuit, convex, and less copper, are detected and identified by using the method of contour contrast. The experiment shows that the proposed method is feasible and the on-line detection of the surface defects of PCB is realized effectively and accurately.