A Video Denoising Method Based on Grouping the Similar Blocks and Surfacelet

Geng Peng ； Xian-Bin Li ； Wan-Hai Yu ； Shi-Qiang Yan ； Jian-Hu Liu

《Research Journal of Applied Sciences, Engineering and Technology》 3卷10期 (2011/10) Pp. 1182-1187

並列摘要

In this study, a novel video denoising method combining similar block matching based on the PDS searching method, grouping these similar block strategy and Surfacelet transform is proposed. Firstly, we apply the SAD block matching criterion and PDS search algorithm which we proposed by searching all frames for blocks which are similar to the currently processed one. In the complex motion case, the method of PDS can effectively decrease the number of search points and search number. Secondly, the matched blocks are put together to stack into some new third dimension sub-video sequences and because of the similarity between them, the data in the video array exists high level of correlation. We apply the surfacelet transform to them and effectively attenuate the noise by solid threshold shrinkage of the coefficients. Finally, inversely transforming the coefficients and obtaining the denoising video according to the obtained locations in the block matching process. This algorithm is obviously better than other 3D Curvlet and 3D wavelet method in the denoising effect and the PSNR is increased about 0.6dB. In terms of visual quality, the proposed method can effectively preserve the video detail, and the trajectory of motion object is very smooth, which is especially suitable to process the video flames with plenty of large area movement object or background change.

並列關鍵字

NDDFB ； PDS ； SAD ； Surfacelet transform ； video denoising