一种改进的小波阈值图像去噪算法

ccording to analysis threshold estimation of wavelet cofficients by the conventional soft-threshold function and hard-threshold function, this paper aimed at the characteristics of the hard-threshold with discontinuous, shortcoming of the permanent bias in soft threshold function and the characteristics of the compromise function with non-derivative. This paper proposes a wavelet threshold de-nosing function which based on a fundation of wavelet continuous, derivative coefficients and variable threshold bias. By the simulated test on MATLAB, results show that, whether form the visual effects, or form signal to noise ratio (SNR) and the root mean square error (RMSE) of view, the improved thresholding de-noising effect is better than the traditional soft and hard threshold and the effect of a compromise de-noising algorithms. At last, this method is also be realized easily!