An Edge Detection Algorithm Based on Multi-resolution Image

In this paper, a multi scale Gaussian edge detector is constructed. According to transfer properties across scales of the wavelet modules of the signal edge and the noise edge, we combine the properties of edges in different scales and propose a multi scale edge fusion algorithm consisting of edge transfer, edge inherit and edge growth. The edge is an important content of obtaining information of a picture, as the most basic characteristic of images. The wavelet transform can detect part mutation, and can do it combining multi scale information, so wavelet has become a good too l of detecting information of edge images. The result of experiments show s that this algorithm can get rid of the affect of noise and the edges fused have precise posit ion and intact contour.