Detecting Defects in Golden Surfaces of Flexible Printed Circuits Using Optimal Gabor Filters

This paper studies the application of advanced computer image processing techniques for solving the problem of automated defect detection for golden surfaces of flexible printed circuits (FPC). A special defect detection scheme based on semi-supervised mechanism is proposed, which consists of an optimal Gabor filter and a smoothing filter. The aim is to automatically discriminate between "known" non-defective background textures and "unknown" defective textures of golden surfaces of FPC. In developing the scheme, the parameters of the optimal Gabor filter are searched with the help of the genetic algorithm based on constrained minimization of a Fisher cost function. The performance of the proposed defect detection scheme is evaluated off-line by using a set of golden images acquired from CCD. The results exhibit accurate defect detection with low false alarms, thus showing the effectiveness and robustness of the proposed scheme.