Automatic Defect Inspection Of PCB Bare Board Based On Machine Vision

Printed circuit board (PCB), as a kind of information carrier that integrates varieties of electronic components, is the basic link of modern electronics industry. Along with the development trend of high precision, multi-layered and miniaturization of PCB, the traditional manual inspection methods, such as eyeballing, electronic detection can not meet the needs of the actual production.In view of the problem existing in PCB bare board detect inspection, this dissertation mainly researched the principle and algorithm for PCB bare board detect inspection based on machine vision. Then the algorithm was is verified and evaluated through experiments. It laid the foundation for the future development of automatic real-time detect inspection system of PCB. The contents and results of the research can be briefly summarized as follows:(1)A detect inspection system of PCB was designed, which based on machine vision. Optical characteristics of PCB materials were analyzed. According to the characteristics and inspection requirements, the image acquisition scheme was determined. The camera and illumination was chosen. Then the experiment platform was established and the images were acquired.(2)The camera model and the lens distortions principle were analyzed. For PCB image the mainly distortion is barrel-shaped. The distortion image could be recovered by polynomial fitting and gray restoration of bilinearity interpolation. The control points were extracted by the checkerboard test images. Then emendation model parameters were calculated and the distorted image was corrected effectively.(3) On the PCB image there symmetrically distributed some vertical reference baselines. First, these baselines were detected rapidly by restricted areas Hough transform and the intersection of the lines were chosen as the feature point. Then the affine registration between the target image and the standard image could be completed. After subtraction, false defects were removed by threshold segmentation and morphological processing in the difference image. The locations of defect areas were obtained.(4) A defect classification algorithm was proposed to identify the type of detect, which based on contour segmentation. The difference image was in the process of dilation, then the coordinate values of each defect area closed contour points could be obtained by boundary detection. The aligned target image was treated with threshold segmentation. In the processed image, the pixel values of the points whose coordinate values had been get above were obtained. According to the analysis of the pixel values and the judgment of whether the defect was lack of material or not, the type of detects could be quickly determined by tree classification.(5) The PCB defect inspection software system was developed. The experimental results on 200 PCB images indicated that the correction rate of detection is 96.5%. The testing time of PCB bare boad whose size is 350mm×250mm was about 49s. The algorithm could basically meet the needs of the real-time detection.