Image Segmentation Method Based On The Measure Of Fuzziness

Selecting correct threshold is a critical issue for image segmentation. The validity of thresholding is very important. This paper is engaged in the traditional bilevel thresholding algorithm. A new thresholding method and detecting edge method are proposed in this paper.In the part, which improves the bilevel thresholding method, we denote a membership function for the "object" in the image, firstly. Here, the "object" is the region which interesting to the study. Secondly, in order to improve the effectiveness, the algorithm utilizes the measure of fuzziness of the image and the conditional entropy to identify the appropriate thresholds automatically. The algorithm is useful to segment the image with multimodal histogram. Using MATLAB, the experiments show that this improved algorithm's validity and pertinence are better.Furthermore, a new detecting edge method based on minimum the measures of fuzziness principle is presented. Thus, on the one hand, it can distinguish the pixels of input image into edge and nonedge effectively due to applying direction information measures; on the other hand, the method divides the pixels into two types through minimizing the measures of fuzziness, such that we obtain the edge of image. The experimental results indicate that the proposed method has good performance in distinguishing and extracting boundaries.