Fractional Order Edge Detection Method

Abstract

Edge detection is a key topic in the field of image processing and computer vision. It plays a crucial role in image segmentation, pattern classification and other work. A new fractional order edge detection method is proposed in this paper. This operator combines CRONE operator and compound derivative operator, and a one-dimensional fractional-order integral filter in a complex derivative operator is converted into a two-dimensional fractional-order integral filter, and a two-dimensional differential mask is constructed by extending the horizontal and vertical components of the CRONE operator differential template. Then, the integral filter and the differential mask are simultaneously used for edge detection. The experimental results show that this operator can extract a target with clear edges and complete structures for all kinds of noise and non-noise pictures. Compared with CRONE operator and composite derivative operator, detection selectivity and noise immunity are improved, and edge enhancement and structure retention are realized.