부산대학교 도서관

In order to expand the range of fractional differential orders, overcome the shortcomings of traditional fractional differentials that assume that surrounding pixels have the same effect on the center pixel during image enhancement, an improved fractional differential image enhancement algorithm is proposed in this paper. First, the similarity factor is calculated according to the distance between the center pixel and the surrounding pixels, and then an improved image fractional differentiation enhancement template is obtained according to the R-L fractional differentiation. Experiments show that, compared with the traditional fractional differentiation method, this method extends the order of differentiation and can better enhance the information of the edge and texture of the image without causing excessive enhancement of the image.