The Study Of Image Denoising Method Based On Wavelet Edge Detection

The real-life images are always corrupted by noises. For the need of further images analysis and communication, the noises need to be reduced in image per-processing. Recently, with the improvement of wavelet theory, wavelet analysis has penetrated into many fields. Wavelet transform has the characteristic of "mathematics microscope", thus it can not only wipe off noises but also retain the original image details. Traditional denoising methods can filter noise, at the cost of the image details being fuzzy. This paper presents a new image denoising method based on edge detection and MAP estimation in order to remove the noises while retaining as many the important signal features as possible.Traditional methods of edge detection are base on one-order derivative's maximum, or two-order derivative's zero-crossing. This kind of edge definition is very sensitive to noises. So edge detection should be carried out in large scale, by smoothing the images. One of the shortcomings of edge detection in large scale is that it's difficult to locate edge precisely, which will make mistakes in pattern recognition in edge features. With multi-scale characterization, wavelet analysis was widely used to multi-scale edge detection. We make good use of the properties of multi-scale edge information to locate image edge with B-Spline wavelet. This can obtain both better image edge and higher accuracyThe detected edge coefficients will be protected from denoising. With sub-band high-frequency components, global thresholding can throw off key formation of image, while local thresholding under the rule of Bayesian maximum probability a posteriori(MAP) with locality is calculated by local components, which provids better performance. The denoising algorithm of this paper is proved by simulation experiments. The proposed method compared to other denoising algorithms improves performance and the quality of subjective visual effects. It is found that image contour is clearer and detail is sharper.The main contents are as following:(1) This paper discussed the developing process of wavelet theory, application of wavelet in image denoising and edge detection and advantages of wave denoising.(2) The paper analyzed application of wavelet in image edge detection, made good use of characteristic of wavelet, got different scale wavelet using B-Spline, picked-up image edge separately in each scale; then used multi-scale characteristic of edge information, fused edge image of each scale, exerted advantage in each scale, got precise single image edge.(3) Aimed at the shortcoming that traditional denoising method couldn't reserve edge characteristic and same threshold "overkills" wavelet coefficient, we present wavelet thresholding denoising method based on edge detection and MAP estimation. This method located image edge by wavelet edge detection which was not affected by threshold before denoising image. Obviously, it was important to select proper threshold function. Simulation comparison was presented by using soft-thresholding.