Based On The Gerber Pcb Defects Of Photoelectric Detection System Software Research

As the rapid development of the PCB (Printed Circuit Board), the quality of PCB production has become an important cornerstone of business survival, so the quality of PCB should be controlled strictly. Because the foreign PCB defect detection systems are very expensive and there are still some shortcomings in domestic products, it is necessary for us to further research.Layers of the PCB have different colors, if all layers of the PCB using the same threshold, the results will have a greater error, so different layers of PCB should be checked differently. The accuracy of traditional method using the characteristics of the PCB image to match and stratify is low. The Gerber files are standard for producing PCB. Each Gerber file corresponds to a layer of PCB, so Gerber files are used for PCB image processing to improve the detection accuracy in this dissertation.The design and implementation for the software part of the optoelectronic inspection system for PCB defects is presented in this dissertation, including the efficient detection algorithms and friendly graphic user interface.In PCB images processing respect, the standard PCB and the test PCB are both matched with the Gerber, so the accuracy for the matching is improved. Then the PCB is stratified by the Gerber, the threshold is calculated by the feature of the local part, so the parts of PCB can be detected differently. Compared with the traditional method by color stratification, this method improves accuracy.In the speed of the detection, the multi-core parallel processing has been used to accelerate the algorithm, and it realizes the truly parallel computing, the resources of the computer can be taken full advantage. In this thesis, the connected component detection algorithm has been taken for example to describe the parallel computing.In order to develop a friendly interactive interface to help user make better use of the system, we use Qt library and wizard mode to design the application software. The two modules in the software system are described in detail in this thesis, including the establishment of the standard library and defect detection module. The most contributions are utilization of Gerber with image processing in the detection system, and implement a set of cross-platform software applications. Experimental results show that the defect detection is very accurate, and the false accept rate is very low, and the checking time of this system is in10s. To our knowledge, the method that uses Gerber to match and stratify PCB image has not been reported in China.