Automatic parameters selection method of edge detector in the unstructured environment

Studies of edge detect algorithms are carried out based on grads in this paper. In order to improve the ability of parameters selection adaptively, a new method of automatic selection of edge detector parameters based on algorithm of Ostu and statistical theory is presented. The global optimized parameters are determined automatically and adaptively through an optimized method of statistics. Canny edge detector is used to prove the validity for proposed method. Experiment results show that, based on proposed method, the noise is effectively reduced, the optimized parameters is automatically selected, and the precision of edge detection is adaptively improved. This kind of detecting method has the advantages to apply to unstructured environments without prior-knowledge.