Improved Canny Algorithm for Edge Detection of Core Image

Abstract

Aiming at the disadvantages of edge detail loss and false edges in conventional Canny algorithm, this paper proposes an optimized Canny edge detection algorithm for edge detection of core image by combining self-adaption smooth filtering and self-adaptive threshold value. The algorithm, which retains the advantages of high signal to noise ratio, accurate positioning, single edge single response and etc in conventional Canny algorithm, has stronger de-noising capability and better edge-detection effect. The experiment results indicate that the improved algorithm can perform selfadaptive edge-detection on the core image and create a clear coherent detected edge of image which can display the edge feature better.