An improved edge detection algorithm based on fuzzy theory

There exists serious distortion in Pal-King edge detection algorithms. In order to solve the problem, an improved method based on fuzzy theory and the maximum between cluster variance is presented by studying Pal.King algorithm synthetically. Firstly, the histogram of an image is calculated and researched. If there is one peak in the image histogram, the Pal-King operator can be used to process the image directly. Otherwise, the fuzzy threshold is set to enhance the image. Secondly, the image is filtered by the Gaussian filter. Then the non-maximum suppression method is applied to locate the edge and process the gradient magnitude. Thirdly, a two-threshold method is used to detect and connect the image edge. At last, a series of experiments have been done to compare this new algorithm with Sobel operators, Pal-King operators and Canny operators. It is shown from the experimental results that the method proposed in this paper is efficient.