PCB defect detection based on pattern matching and segmentation algorithm

Abstract: This project presents a new implementation which separates two of the existing groups containing two defects each into four new groups containing one defect each by processing synthetic images of bare through-hole single layer PCBs using MATLAB Image Processing. This also will use Mathematical Morphological Toolbox here. Morphological process involves techniques such as dilation, erosion, opening and closing which helps in partitioning the images and associates certain types of defects with certain patterns. This project separates the defects in larger groups into smaller groups. This increases the efficiency of the inspection system in classifying defects. Since certain PCB pattern are produced in different processes, classification of defects can help in determining the root causes of error and reduce production cost in the long run. Focus is given to separate groups of defects in the hole segment, thick line segment and thin line segment.

Keywords: Image Processing, Printed Circuit Board, Defect Classification.