Modified self-adaptable wavelet transform arithmetic in image edge detection

In order to solve the problem between suppressing noise and accuracy of the edge, this paper gave a self-adaptable scale selection method with local entropy. It adopted the local entropy which could divide image precisely to preprocess the image. The self-adaptable method of wavelet transform selected the scale of each region. Each region used corresponding scale of wavelet transform to get image edge. Experimental results show that the proposed method provides clearer and more accurate edges than wavelet transform self-adaptable arithmetic based on mean square deviation.