基于Zernike矩的亚像素边缘检测算法

im to the drawbacks of traditional algorithms that have lower location precision and detect wider edge, the sub-pixel edge detection operator which based on Zernike moments were proposed in this paper. Moreover, a mask of size seven by seven was caculated. The algorithm is in the form of integral operator which is not sensitive to noise, resulting in effectively suppresses noise, detects the better edge of image and achieves higher localization. It can overcome the disadvantages of traditional algorithms such as Sobel operator that detects the edge in thicker, larger error and is more sensitive to noise. Compared to Sobel operator, the results of the experiment verify that the Zernike moments can precise positioning. Moreover, the accuracy can reach the level of sub-pixel.