标题:Image denoising in hybrid frequency domain
作者:Liu, Yushu; Jiang, Mingyan
通讯作者:Liu, YS
作者机构:[Liu, Yushu; Jiang, Mingyan] Shandong Univ, Sch Informat Sci & Engn, Jinan, Peoples R China.
会议名称:International Conference of Information Science and Management Engineering (ISME)
会议日期:2013
来源:INFORMATION SCIENCE AND MANAGEMENT ENGINEERING, VOLS 1-3
出版年:2014
卷:46
页码:651-657
DOI:10.2495/ISME20130841
关键词:Image denoising; Gabor; contourlet; Wiener filtering
摘要:Image denoising is a classic problem in image analysis, and the denoising result will affect the subsequent image processing, such as feature extraction. In this paper, we proposed a new denoising method based on Gabor transform and contourlet transform. From the Gabor transform we can keep the periodic and smoothed texture information of the image, and the directional and suddenly changed information can be preserved by contourlet transform. The experiment results indicate that our method is simple, intuitive, and effective. Compared with the method locally adaptive window-based denoising using maximum a posterior (LAWMAP), PSNR gained by our method is higher by 0.7 dB at most, and visual quality is also improved.
收录类别:CPCI-S
资源类型:会议论文
TOP