Image edge detection based on wavelet transform and Canny operator

Abstract

A method for image edge detection based on image fusion is presented in this paper. Since traditional wavelet transforms are unable to control the noise well and the edge is not consistent with direction properties, some improvements are made and a new kind of wavelet transform is proposed. It detects the edge of original image by means of new wavelet transform and Canny operator respectively, then produces a new image by fusing and analyzing the two results based on the experimental results. It is shown that the proposed method provides clearer and smoother edges than that using Sobel or wavelet transformation algorithms alone. This algorithm is simple, useful and easy to implement.