DocumentCode
3272798
Title
Image Denoising Based on Combination of Wiener Filter and Wavelet Shrinkage
Author
Dai, Wenzhan ; Ye, Yulei
Author_Institution
Zhejiang Gongshang Univ., Hangzhou
fYear
2007
fDate
20-24 March 2007
Firstpage
421
Lastpage
425
Abstract
First, a new thresholding function based on the wavelet shrinkage is presented. The shortcoming of discontinues function in hard thresholding and of the permanent bias in soft thresholding can be overcome. And then a new image de-noising approach based on combination of Wiener filter and wavelet shrinkage is also put forward. In this new approach, First, a new thresholding function based on the wavelet shrinkage is presented. The shortcoming of discontinues function in hard thresholding and of the permanent bias in soft thresholding can be overcome. And then a new image de-noising approach based on combination of Wiener filter and wavelet shrinkage is also put forward. In this new approach, the image noise is filtered out while the edges of image are well preserved. Simulation result shows that the new approach presented in this paper is more effective and predominant than other de-noising methods.he image noise is filtered out while the edges of image are well preserved. Simulation result shows that the new approach presented in this paper is more effective and predominant than other de-noising methods.
Keywords
Wiener filters; image denoising; wavelet transforms; Wiener filter; image denoising; thresholding function; wavelet shrinkage; Continuous wavelet transforms; Discrete Fourier transforms; Discrete wavelet transforms; Educational institutions; Fourier transforms; Image denoising; Nonlinear filters; Spatial filters; Wavelet domain; Wiener filter; Image de-noising; Wavelet shrinkage; Wiener filter;
fLanguage
English
Publisher
ieee
Conference_Titel
Integration Technology, 2007. ICIT '07. IEEE International Conference on
Conference_Location
Shenzhen
Print_ISBN
1-4244-1092-4
Electronic_ISBN
1-4244-1092-4
Type
conf
DOI
10.1109/ICITECHNOLOGY.2007.4290510
Filename
4290510
Link To Document