Abstract

Study on an new algorithm of image contour extracting

Author(s): Shaofei Wu

An algorithm based on human-computer interaction is put forward aiming at the disadvantages of low accuracy of image contour and low degree thinning of contour. First, to reduce image noise is necessary by filtering, the following thing is to change image into gray image. The gray image is segmented by fuzzy c-means clustering algorithm, In the algorithm, the edge is traversed at given initial vertex. According to neighbor relationship between the point on the contour with neighbor pixels, curve can be confirmed to remain or not. In the following, curve contour vertices can be concluded by the traversal order. Experimental results show that high contour accuracy can be obtained to detect image by the algorithm. In regional consistency, the algorithm is better than some traditional algorithms of image edge detection