The Algorithm of Image Denoising Based on the Improved Method of Wavelet Thresholding
Wang Qion · Techniques of Automation and Applications · 2013
Image denoising is the most basic and important preliminary in work image processing, A threshold function of Garrote based on the attenuation law is proposed, and a wavelet thresholding method based on the improved threshold function is used in image denoising process. At last, the effectiveness of the algorithm improved by this paper is verified through the MATLAB simulation experiments. With analyzing the influences on the effect of image denoising by the method of wavelet threshold, and for the problem of a certain amount of image detail information is lost while the noise is removed by the denoising algorithm. And the shortcomings that the wavelet coefficients below the threshold is set to zero blindly without considering the wavelet coefficients below the threshold may contain the image detail information are improved. In order to retain the more image detail information, a attenuation method is taken in below a threshold value of the wavelet coefficients of Garrote threshold function. The experiments show that the noise can be effectively removed and a large number of image edges and details can be retained by the denoising algorithm based on the improved wavelet threshold.