Image edge detection based on statistical features

摘要

Classical edge detection method has great limitation, because the noise contained in the image has significant influence on the results, while the speed of edge detection and whether the edge can be detected or not are also the concerned problems. This paper proposed two edge detection methods based on the statistical features, which can accurately detect the edges and suppress the impact of the noise on the results, while the edge has a good consistency. Extensive testing results have shown great potential on our novel method.