Image Denoising with Non-Local Means in the Shearlet Domain

Author

Zhang, Xiaohua ; Zhang, Qiang ; Jiao, L.C.

Author_Institution

Key Lab. of Intell. Perception & Image Understanding of the Minist. of Educ. of China, Xidian Univ., Xi´´an, China

fYear

2011

fDate

10-12 Jan. 2011

Firstpage

1

Lastpage

5

Abstract

In this paper, a novel method for image denoising is proposed which adopts multiscale geometry tool. Firstly the image is decomposed by discrete shearlet transform. The shearlet coefficients of each direction approach the generalized Gaussian distribution. We use the principal component analysis (PCA) for every similarity window of shearlet coefficients. Then we use Generalized Gaussian model of non-local means method to handle the shearlet coefficients. Finally, we reconstruct image with the new shearlet coefficients to obtain the result. Numerical results show that our algorithm competes favorably with nonlocal means algorithms in the case of high noise.

Keywords

Gaussian distribution; discrete transforms; image denoising; image reconstruction; principal component analysis; discrete shearlet transform; generalized Gaussian distribution; generalized Gaussian model; image denoising; image reconstruction; non local means; principal component analysis; shearlet coefficients;

fLanguage

English

Publisher

ieee

Conference_Titel

Multi-Platform/Multi-Sensor Remote Sensing and Mapping (M2RSM), 2011 International Workshop on

Conference_Location

Xiamen

Print_ISBN

978-1-4244-9402-6

Type

conf

DOI

10.1109/M2RSM.2011.5697407

Filename

5697407