Share Email Print

Proceedings Paper

A new adaptive algorithm for image denoising based on curvelet transform
Author(s): Musheng Chen; Zhishan Cai
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The purpose of this paper is to study a method of denoising images corrupted with additive white Gaussian noise. In this paper, the application of the time invariant discrete curvelet transform for noise reduction is considered. In curvelet transform, the frame elements are indexed by scale, orientation and location parameters. It is designed to represent edges and the singularities along curved paths more efficiently than the wavelet transform. Therefore, curvelet transform can get better results than wavelet method in image denoising. In general, image denoising imposes a compromise between noise reduction and preserving significant image details. To achieve a good performance in this respect, an efficient and adaptive image denoising method based on curvelet transform is presented in this paper. Firstly, the noisy image is decomposed into many levels to obtain different frequency sub-bands by curvelet transform. Secondly, efficient and adaptive threshold estimation based on generalized Gaussian distribution modeling of sub-band coefficients is used to remove the noisy coefficients. The choice of the threshold estimation is carried out by analyzing the standard deviation and threshold. Ultimately, invert the multi-scale decomposition to reconstruct the denoised image. Here, to prove the performance of the proposed method, the results are compared with other existent algorithms such as hard and soft threshold based on wavelet. The simulation results on several testing images indicate that the proposed method outperforms the other methods in peak signal to noise ratio and keeps better visual in edges information reservation as well. The results also suggest that curvelet transform can achieve a better performance than the wavelet transform in image denoising.

Paper Details

Date Published: 26 October 2013
PDF: 5 pages
Proc. SPIE 8921, MIPPR 2013: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications, 89211N (26 October 2013); doi: 10.1117/12.2032223
Show Author Affiliations
Musheng Chen, Quanzhou Normal Univ. (China)
Zhishan Cai, Quanzhou Normal Univ. (China)

Published in SPIE Proceedings Vol. 8921:
MIPPR 2013: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications
Jinwen Tian; Jie Ma, Editor(s)

© SPIE. Terms of Use
Back to Top