Adaptive Image Edge Detection Algorithm Based on Canny Operator

Abstract

For single edge detection methods causing important and weak gradient change edge missing problems, this paper adopts the method of combining global with local edge detection to extract edge. The global edge detection can obtain the whole edge, which uses adaptive smooth filter algorithm based on Canny operator. Compared with effect of edge detection from the Canny operator and Sobel operator, the edge from improved Canny operator is the most complete and rich, do not contain false edge. To the whole detection failed to get the edge, the paper selects local area detection method for edge extraction. Local edge detection which uses distance weighted average method based on k-average method can overcome the impact of outliers on clustering effectively. Complete skull image edge is got through edge detection method that combines global with local. Compared with the Canny edge detection method, this algorithm can extract image edge effectively, and have the powerful anti-noise ability.