Image edge detection method based on the direction feature of fuzzy entropy

To improve the image edge detection capability, this paper presents an image edge detection method based on the direction feature of fuzzy entropy. Firstly, the proposed method operates the fuzzy membership function and fuzzy entropy function for achieving a neighborhood fuzzy entropy feature space from the image feature to enhance the fuzzy edge contrast, and defines twelve valid edge direction structures in the 3×3 neighborhood of the feature space. Then it extracts the valid direction structural information arrays of every pixel point, and combines the direction structure measurement arrays to make non-maxima suppression. Finally the proposed method implements an adaptive threshold to estimate and determine image edges. Experimental results illustrate that the proposed method has better performance in the image edge detection with more image detail information.