| Image information is one of our the most important source of information, but the image information is very easy in the acquisition and transmission by the outside noise pollution, so that some details of the features of the image can not be effectively identified, leading to the subsequent processing image is affected by. Therefore, image denoising is the follow-up image processing such as an important step in the image compression, image detection and feature extraction, and has the very strong practical significance and application value.This paper based on the wavelet theory, focusing on the wavelet threshold denoising method for a more in-depth analysis and research, the main content of this paper has the following several aspects:(1) This paper presents the principle of wavelet transform for image denoising, introduces three kinds of wavelet transform for image denoising methods, focus on the wavelet image threshold denosing methods to select the threshold function of analysis and research, an improved threshold function is proposed, and the improved threshold function is applied to wavelet image threshold denoising in wavelet image threshold, at the same time, and several traditional denoising methods were compared and experimental simulation, and from the two aspects of subjective and objective to analyze and compare the denosing effect.(2) The analysis and study of median filter and its improved method, through simulation comparison, the adaptive median filter method to verify the validity and superiority of the impulsive noise of image denoising.(3) For the mixed noise of impulse noise and Gauss noise image denoising, this paper proposed combined median filtering with wavelet transform, firstly using adaptive median filter removing impulse noise, then using the improvement of wavelet threshold image denoising removing the Gauss noise. Through the simulation experiment, the proposed method is better than the single median filter or wavelet threshold denoising. |
PDF Full Text Request