Sub-pixel edge detection method based on Zernike moment

This paper presents an image sub-pixel edge detection method based on Zernike moment. Firstly, we offered the edge model parameters calculation method through Zernike moment definition and the rotation invariance. Then, the orthogonal complex polynomial from 0 order to 2 order of the Zernike moment was calculated, in order to save time we calculate the image Zernike moment mask values in advance. Meanwhile, to improve the Zernike moment edge detection operator´s positioning accuracy and suppressing noise abilities we use 9×9 grid mask. Finally, we applied these Zernike moment edge detection operators for some standard facial recognition. The experiment results show that this method has higher detection precision compared with traditional methods.