Fuzzy Edge Detection Method

Edges of an image reflect the information of the image mostly. They contain the basic characteristics. Edge detection is a vital part of many image processing and pattern recognition systems. Typical areas of applications are image segmentation, stereo vision, and identification of objects as in automatics target recognition . It is one of the most parts in image processing. The traditional edge detection methods can be divided into two categories, time domain and frequency domain. These methods are not very effective to the images which edges are fuzzy. In this paper, an edge detection method based on fuzzy reasoning is improved. To extract edges effectively , edge pixels are classified according to the difference of intensity distribution, and several effective edge features are defined. It uses the different edge features to make the fuzzy rules and applys fuzzy reasoning(min-max barycenter method)to detect the edges. Finally, in this paper, the results are given applying this method to process the images . And the comparisions are made with some traditional edge detection methods. It is proved that the results applying the edge detection method based on fuzzy reasoning are better than those applying the traditional edge detection methods. It indicates that the method improved in this paper is more effective to detect edges than the traditional methods.