A noise elimination procedure for wavelet-based printed circuit board inspection system

Image difference operation is frequently used in automated printed circuit board (PCB) inspection system as well as in many other image processing applications. The previously proposed wavelet-based PCB inspection approach also incorporated the point-to-point image difference operation as part of its algorithm. Unfortunately, during the implementation, this operation brings along the unwanted noise due to misalignment and uneven binarization. Thus, this paper proposes a method to eliminate, if possible, or to reduce as much as possible such noise during the computation of defect detection. The results of applying the proposed method showed a significant improvement during the real-time inspection of printed circuit boards.