Research On Defect Inspection Method Of Circuit Board Based On Image Processing

With the development of integrated circuit technology, the size of electronic components is becoming smaller. There may be some defects inevitably during the component jointing, such as component missing and component mischoice. In the engineering application and national defense area, with the intelligentization and imformationization, the function of instrument and weapon systems becomes rich and more complex, so, much more circuit boards are needed. However, the circuit boards with defects may effect the performance of weapon systems, and may cause collapse of weapon systems which is very serious. It is essential to inspect defect of circuit boards before they come into service. But traditional defect inspection by human has low efficiency, and low stability which effected by visual fatigue. Hence, the research on defect inspection of circuit board based on image processing, is of great practical engineering significance of ensuring circuit board quality of weapon systems, and the research on feature extraction of components and algorithm is also of academic significance to a certain extent.With sufficient investigation and analysis about the research situation of circuit board defect inspection, this paper gets circuit board images through digital camera, and component features on the target circuit board are extracted, then component features compare with standard component features which are extracted from standard circuit board, thus, component defects on target circuit board are inspected. The main research works of this paper are as follows:①Based on the summary of defect inspection actuality and the study on digital image processing, an overall scheme of circuit board defect inspection is proposed.②According to that there are outline of components and some other characters on PCB silkscreen layer, algorithms are proposed to extract component outlines, namely, the sketchy position of every component is extracted.③On the standard circuit board, different feature extraction algorithms are designed to extract different component features, and component standard feature database is established.④According to the characteristic of circuit board defects, a rational defect inspection flow is proposed, different defect inspection algorithms are designed to inspect different defects.⑤Defect inspection experiments are executed with different circuit boards, defect inspection result is analyzed combined with experimental data.Circuit board defect inspection results show that the proposed inspection scheme is rational, and compared with traditional inspection method, defect inspection based on image processing is more efficient and stable.