We developed a scale multiplication-based scheme to improve the performance of traditional Canny edge detector. Taking the advantage of similarities in the filter’s responses at adjacent scales, the new scheme multiplies the responses to enhance edge structures while diluting noise and detect the edges as the local maxima in the scale products. Our theoretical analyses show that scale multiplication can improve the edge localization accuracy and then yield better edge detection results. Experiments on synthetic and natural images were made to test the proposed method.
Bao, P., Zhang, L., & Wu, X. (2005). Canny edge detection enhancement by scale multiplication. IEEE transactions on pattern analysis and machine intelligence, 27(9), 1485-1490.