Abstract: In edge detection, matching with a template derived from the input data is meaningful since it provides an independent measure of confidence in the presence of the employededge model. In this paper, by making good use of the characteristics of wavelet transform and multi scale of edge information, the cubic B spline smoothing filter operator is designed and combined with the embedded confidence to detect edges. Furthermore, the edge detection method for those images badly contaminated by noises is studied, according to the sensitivity of human visual system. The experimental results show that the proposed method can detect the edges of images accurately and efficiently.