The Study On Images De-noising Based On Wavelet Transform And Convolution Morphology

With the development of the technology, images processing has been more and more important in the field of information processing. When images are transmitted, the technical problems also have been paid more and more attention to .Images de-noising is one of important task of images processing. The aim of images de-noising will wipe off the noise, which can more deal with images.In fact , the traditional images de-noising arithmetic is a low-pass filter, which can filtering out noise in the high-frequency information, and also can loss edge, contour and texture information in the high-frequency detail.To present a arithmetic based on Wavelet Transform and Convolution Morphology for images de-noising. Because of the nature of multi-analysis and multi-scale, Wavelet Transform can analyze the images to multi-scale and multi-analysis, disposed by Convolution Morphology and recombined. The typical image with noise has been analyzed, and proved the arithmetic feasibility and availability.We tested and simulated for the standard image at the different noise background, and also contrasted with Wavelet de-noising and Convolution Morphology de-noising. The result has indicated the arithmetic brought down by the paper, which cut down the mean square error (MSE) compared with Wavelet de-noising and Convolution Morphology de-noising, and also more improved the signal noise ratio (SNR) and peak signal noise ratio (PSNR). And the result can more legibly reserve the edge information of the image, which proves the arithmetic feasibility and validity.