Research On Denoising Algorithm Of Gauss Noise In Digital Image

Denoising of digital image is a key step in image preprocessing,which directly affects the results of subsequent image processing,such as image enhancement,image segmentation and image coding.Denoising has important applications in numerous fields of science and engineering,which is to used to suppress noise from degraded images and restore images as much as possible to make the visual quality of restored images much closer to the original ones.Gauss noise is a typical noise type in image,and the research of Gauss noise filtering has always been a hot topic for domestic and foreign scholars.Some typical denoising algorithms were put forward such as mean filtering,Gauss filtering,bilateral filtering,Wiener filtering,total variation denoising,wavelet denoising,nonlocal means filter and so on,which have achieved the purpose of suppressing Gauss noise in the image to some degree.Based on the deeply studying and summarizing of the existing Gauss noise filtering algorithms,some improved typical denoising algorithms are presented in this paper.(1)In the algorithm of wavelet threshold denoising,a new exponential threshold function is proposed to shrink the wavelet coefficients,which overcomes the inherent defects of the traditional soft and hard threshold functions,and the reconstructed signals have better continuity.(2)In regard to the high time complexity and large amount of calculation in the nonlocal means filtering algorithm,the integral graph acceleration technique is applied to the computation of Euclidean distance between similar blocks,which greatly improves the computational efficiency,significantly reduces the running time,and significantly improves the quality of denoised image compared to the original algorithm.(3)A Gauss noise filtering algorithm based on Canny edge detection is proposed.Firstly,the edge of the image is detected by using Canny operator,then an expansion processing is carried out on the edge region.The noise of the edge region is removed by the nonlocal means filter algorithm,and the noise of the non-edge region is removed by the wavelet threshold denoising algorithm.Finally,the two parts of denoising results are merged to produce the output image.The experimental results show that the denoising algorithm based on edge detection has a better denoising performance for the image with little noise intensity compared to other denoising algorithm.The research results of denoising of image Gauss noise can be used for consumer electronics.