Feature Enhancement of Noisy Image by Dyadic Wavelet Analysis

Abstract

The image de-noising method is proposed by multiresolution representations of the dyadic wavelet transformation. Firstly, the original image are decomposed using dyadic wavelet transformation, then the image can be de-noised by way of adopting different thresholds according to the different scales of wavelet coefficients, and finally nonlinear enhancing function are applied for wavelet coefficients with the enhancement method in the corresponding scales. Experiments of the medical images show that compared to the image enhancement method, our approach improves the information entropy and contrast improvement index of the images, enhances edge texture of images.