A self-adaptive edge detection method based on LoG algorithm

Abstract

The capability of edge detection using LoG operator is analyzed in this paper and the paper proposes the deficiency of the LoG operator in the practical application. In order to avoid the defects of the LoG operator, the paper based on the date of experimentation and the Matlab software elicits the relation between the entropy of the gray level co-occurrence matrix and the Gassian space coefficient. Optimum Gassian space coefficient of LoG operator can be self-adaptive acquired based on the entropy of the concrete image. So the methods improved in the paper not only effectively control the most noise of the image, but also locate the edge accurately.