Image Noise Reduction Based On Wavelet Adaptive Threshold Combined With Bilateral Filtering

In the process of acquisition,transmission and conversion of digital image,the image will be polluted by noise due to the mechanical movement of equipment,internal circuit of system and material of equipment.It can reduce the image quality and cause trouble to the subsequent image processing.Wavelet threshold method and bilateral filtering are the main techniques for image denoising.However,the discontinuity of hard threshold function results in Pseudo-Gibbs phenomenon after image reconstruction.There is always a constant deviation between the estimated wavelet coefficients and the original wavelet coefficients of the soft threshold function,which affects the accuracy of image reconstruction.The global threshold of wavelet transform is invariable,which leads to over-strangling of signals.In addition,the traditional bilateral filtering in a filtering window,all the pixels including irrelevant neighborhood pixels participate in the filtering,which is not conducive to the central pixel.In view of the above limitations,the main work of this paper is as follows.Firstly,the traditional wavelet threshold function is improved,the improved wavelet threshold function is continuous in the definition domain,and there is no constant deviation between the estimated wavelet coefficients and the origin al wavelet coefficients.An adjustable parameter is included in the improved wavelet threshold function.The mean square error,peak signal-to-noise ratio and structural similarity of the parameters at different values are obtained by experimental simulation.By analyzing the experimental data,the optimal value of adjustable parameters in the improved wavelet threshold function is obtained.Secondly,the invariant traditional global threshold is improved,and a logarithmic function is introduced on the original basis.The improved wavelet adaptive threshold can be slowly reduced with the step-by-step increase of the wavelet decomposition scale to adapt to the phenomenon that the amplitude of noise coefficient decreases with the increase of the wavelet decomposition scale,so as to avoid the over-strangling of the effective signal.Finally,the gray filtering kernel function of the traditional bilateral filtering is improved,and a filtering threshold is constructed,which is improved to a piecewise function.The advantage of this method is that the neighborhood pixels with large gray difference from the central pixel can not participate in the filtering at all,and the influence of the neighborhood pixels independent of the central pixel in a filtering window can be reduced,and the ability of bilateral filtering to protect the edge of the image target can be improved.The simulation results show that the combination of bilateral filtering and wavelet adaptive threshold function is superior to the traditional denoising algorithm in terms of visual effect,mean square error,peak signal-to-noise ratio and structural similarity.