一种能保留图像边缘信息的去噪新方法  

New Method for Images Denoising while Keeping Edge Information

作　　者：张春梅 张太镒[1] 陆从德[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049

出　　处：《中国图象图形学报》2006年第10期1464-1467,共4页Journal of Image and Graphics

基　　金：国家自然科学基金项目(50276047)

摘　　要：图像处理的目标是消除噪声的同时能保留图像所固有的信息。针对保留边缘信息有效去噪问题,提出了双密度双树复数小波变换的图像去噪方法,该方法综合了双密度小波、双树小波和复数小波的优点,具有更好的方向性,将双树复数小波的6个方向,提高到12个方向,并采用了自适应软阈值对小波变换的系数进行处理,消除图像干扰噪声。本文对加噪图像进行去噪仿真试验,并进一步进行边缘检测,仿真试验结果表明,该方法能有效消除图像噪声并保留图像原有边缘信息,与双密度双树小波相比,去噪效果明显改善,均方误差减小了2.4%。The objective of image processing is to denoise while keep the intrinsic imforrnation. A method is proposed for images denoising with double density dual-tree complex wavelet transform. It integrates the advantages of double density wavelet, dual tree wavelet and complex wavelet. It has improved directionality which increased the directions from 6 to 12. The adaptive soft threshold is applied to wavelet coefficients. It could ameliorate the ability of image denoising. The denoising and edge detection for a typical image is carried out. Simulated experiments demonstrate that the proposed approach is effective on image denoising and can keep the edge information at same time. Its mean square deviation is reduced 2.4% compared with the dual-tree complex wavelet.

关 键 词：双密度双树复数小波变换 去噪 边缘检测 

分 类 号：TN919.8[电子电信—通信与信息系统]