Application of improved Canny algorithm on the IC Chip pin inspection

摘要

Machine vision is widely used in automation detection of chip pins' defect, the accuracy of image edge detection has a very big effect for inspection result. Canny algorithm is a kind of edge detection algorithm which has good comprehensive evaluation, but the Gaussian filter algorithm it used may cause image too smooth and edge fuzzy, and it is very sensitive for impulse noise. This paper proposes an improved switch median filter algorithm, which is applied into Canny algorithm, makes the edge of IC Chip pin more complete and remove noise better. Through experiments, it proved that the method is feasible and effective.