scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

A new adaptive algorithm for image denoising based on curvelet transform

26 Oct 2013-Vol. 8921, pp 399-403
TL;DR: The simulation results indicate that the proposed method outperforms the other methods in peak signal to noise ratio and keeps better visual in edges information reservation and the results suggest that curvelet transform can achieve a better performance than the wavelet transform in image denoising.
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.
References
More filters
Proceedings ArticleDOI
07 Oct 2001
TL;DR: In the tests reported here, simple thresholding of the curvelet coefficients is very competitive with 'state of the art' techniques based on wavelets, including thresholded of decimated or undecimated wavelet transforms and also including tree-based Bayesian posterior mean methods.
Abstract: Summary form only given, as follows. We present approximate digital implementations of two new mathematical transforms, namely, the ridgelet transform and the curvelet transform. Our implementations offer exact reconstruction, stability against perturbations, ease of implementation, and low computational complexity. We apply these digital transforms to the denoising of some standard images embedded in white noise. In the tests reported here, simple thresholding of the curvelet coefficients is very competitive with 'state of the art' techniques based on wavelets, including thresholding of decimated or undecimated wavelet transforms and also including tree-based Bayesian posterior mean methods. Moreover, the curvelet reconstructions exhibit higher perceptual quality than wavelet-based reconstructions, offering visually sharper images and, in particular, higher quality recovery of edges and of faint linear and curvilinear features.

857 citations


"A new adaptive algorithm for image ..." refers background in this paper

  • ...The curvelet decomposition is the sequence of the following steps [5][6][7] : 1....

    [...]

Journal ArticleDOI
TL;DR: Numerical results show that the proposed algorithm can obtained higher peak signal to noise ratio (PSNR) than wavelet based denoising algorithms using MR Images in the presence of AWGN.
Abstract: Image denoising has become an essential exercise in medical imaging especially the Magnetic Resonance Imaging (MRI). This paper proposes a medical image denoising algorithm using contourlet transform. Numerical results show that the proposed algorithm can obtained higher peak signal to noise ratio (PSNR) than wavelet based denoising algorithms using MR Images in the presence of AWGN.

46 citations

Journal ArticleDOI
TL;DR: A novel denoising method is presented that outperforms its wavelet-based counterpart and pro- duces results that are close to those of state-of-the-art denoisers.
Abstract: We perform a statistical analysis of curvelet coefficients, distinguishing between two classes of coefficients: those that contain a significant noise-free component, which we call the “signal of interest,” and those that do not. By investigating the marginal statistics, we develop a prior model for curvelet coefficients. The analysis of the joint intra- and inter-band statistics enables us to develop an appropriate local spatial activity indicator for curvelets. Finally, based on our findings, we present a novel denoising method, inspired by a recent wavelet domain method called ProbShrink. The new method outperforms its wavelet-based counterpart and produces results that are close to those of state-of-the-art denoisers.

36 citations


"A new adaptive algorithm for image ..." refers background in this paper

  • ...[6]Linda Tessens,Alin Alecu,Wilfried Philips....

    [...]

  • ...The curvelet decomposition is the sequence of the following steps [5][6][7] : 1....

    [...]

01 Jan 2010
TL;DR: An adaptive threshold method based on curvelet transform to estimate noise and remove it from digital images in order to achieve a good performance in this respect is proposed.
Abstract: Image Denoising has remained a fundamental problem in the field of image processing. This paper proposes an adaptive threshold method for image denoising based on curvelet transform to estimate noise and remove it from digital images in order to achieve a good performance in this respect. The proposed adaptive threshold method is more efficient in estimate and reduce noise from images which have random, salt & pepper and Gaussian noise. Experimental results show that the proposed method demonstrates an improved denoising performance over related earlier techniques according to increasing of PSNR values of enhanced images by 0.044 at Random,1.05 at salt& pepper and 0.457 at Gaussian noise.

21 citations


"A new adaptive algorithm for image ..." refers background in this paper

  • ...Therefore curvelet transform is better than wavelet in the expression of image edges such as the geometry characteristic of curve [4] ....

    [...]

  • ...Engineering letters, 14:2, EL_14_2_16 (Advance online publication:16 May 2007) [4]Aliaa A....

    [...]