Edge extraction technique using the accumulation of self-reinforcement of search trajectories of independent-edges

Edge extraction is an important stage to image understanding. To improve past methods for edge extraction, this paper presents an algorithm of accumulation of self-reinforcement of independent edges. First, use repetitive random searches to obtain various possible independent-edge trajectories in edge images. Then, self-reinforce and accumulate the search trajectories respectively. At last, extract the edges according to the results of the accumulation of self-reinforcement. Through lots of experiments and comparing with other techniques of edge extraction: relaxation algorithm and heuristic search algorithm, this method can extract edges effectively and suppress the noise significantly.