Adaptive Image Edge Detection Model Using Improved Canny Algorithm

Abstract

Canny algorithm is one of the most widely used edge detection methods based on the optimal thought. However, it still has some drawbacks. In this paper on adaptive edge detection model based on improved Canny algorithm is proposed. Firstly, we replace the Gaussian smooth in standard Canny algorithm by the proposed morphology method to highlight the edge information and reduce the noise; secondly, the fractional differential theory is utilized to calculate gradient value, which further eliminate noise and enhance image details; next, we propose an interpolation method for non-maximum suppression, leading to a more accurate edge location; finally, a method based on Otsu's threshold method is proposed to get adaptive threshold. Compared with Canny algorithm and other existing methods, the proposed method has better detection accuracy and robustness.