Two-stage image denoising by principal component analysis with local pixel grouping

Zhang, Lei; Dong, Weisheng; Shi, Guangming

doi:10.1016/j.patcog.2009.09.023

Cited by 559 publications

(398 citation statements)

References 26 publications

Supporting

Mentioning

396

Contrasting

Unclassified

Order By: Relevance

“…In this paper we present a novel method related with the filter proposed by Zhang et al (2010) where similar patches are grouped together prior performing the PCA decomposition. Differently from Zhang et al (2010) (which was developed for natural images with stationary Gaussian noise) we group similar patches using a pre-filtered guide image that improves the group selection process in noisy conditions yielding in a sparser group definition.…”

Section: Non-local Pca Denoisingmentioning

confidence: 99%

“…Differently from Zhang et al (2010) (which was developed for natural images with stationary Gaussian noise) we group similar patches using a pre-filtered guide image that improves the group selection process in noisy conditions yielding in a sparser group definition. Besides, we have made our proposed method fully adaptive by internally estimating the local amount of noise within each group of patches.…”

Section: Non-local Pca Denoisingmentioning

confidence: 99%

See 1 more Smart Citation

MRI noise estimation and denoising using non-local PCA

Manjón

Coupé²,

Buades³

2015

Medical Image Analysis

145

122

View full text Add to dashboard Cite

This paper proposes a novel method for MRI denoising that exploits both the sparseness and self-similarity properties of the MR images. The proposed method is a two-stage approach that first filters the noisy image using a non local PCA thresholding strategy by automatically estimating the local noise level present in the image and second uses this filtered image as a guide image within a rotationally invariant non-local means filter. The proposed method internally estimates the amount of local noise presents in the images that enables applying it automatically to images with spatially varying noise levels and also corrects the Rician noise induced bias locally. The proposed approach has been compared with related state-of-the-art methods showing competitive results in all the studied cases.3

show abstract

Section: Non-local Pca Denoisingmentioning

confidence: 99%

Section: Non-local Pca Denoisingmentioning

confidence: 99%

MRI noise estimation and denoising using non-local PCA

Manjón

Coupé²,

Buades³

2015

Medical Image Analysis

145

122

View full text Add to dashboard Cite

show abstract

“…In comparison to exemplar-based approaches for image modeling [14], [19], we propose an unsupervised method that uses no library of image patches and no computational intensive training algorithms [23]. Our adaptive smoothing introduces the joint spatialrange domain as the non-local means filter [13] but has a more dominant adaptation to the local structure of the data therefore the size of windows and control parameters are predicted from local image statistics as follows.…”

Section: Problem Statementmentioning

confidence: 99%

A New Approach to Image Denoising by Patch-Based Algorithm

Rajanesh¹,

Kollem²

2016

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

Different types of denoising methods are existed in databases. But every method has its own uniqueness. We propose a new method that is adaptive patch based system for image denoising. The approach depends on a pointwise selection of narrow image patches of precise size in the variable neighborhood of each pixel. Our contribution is to engage in each pixel the weighted sum of data points not outside an adaptive neighborhood, in a sense that it balances the efficiency of estimation and the stochastic error, at each contiguous position. In this paper introducing spatial adaptivity, we prolong the work that can be designed as a development of bilateral filtering to image patches. This method is tested by using AWGN and images are taken from databases with different resolutions. The performance of the denoising system is computed in terms of PSNR by taking different noise levels.

show abstract

“…Spatial domain filters like lee and map filters [1][2][3][4][5][12][13] gave better despeckling results and are failed in preserving the edge details. Wavelet domain filters [6][7][8][9] have produced better response than spatial filters. That is the reason, the researchers have concentrated on transform domain filters [14][15][16]18,20].…”

Section: Introductionmentioning

confidence: 99%

RISAT-1 Image Despeckling using a Modified Undecimated BM3D Technique

Y¹,

Radhika²

2016

IJIGSP

View full text Add to dashboard Cite

Abstract-In synthetic aperture radar (SAR) imaging, the transmitted microwave pulses from space born antenna interacts with ground objects and returned energy or back scattered energy will be collected to get backscattered image. In SAR image processing, a not anticipated noise (speckle noise) is added due to the coherent imaging system, which makes the image analysis troublesome. For better SAR image processing, the noise is to be removed or minimized in the begging stages of pre-processing and texture features are to be effectively maintained. The wavelet based Block Matching 3D (BM3D) method is normally considered as the state of art technique in the area of denoising of images. This method generally depends on up and down sampling conversion. In this paper, it is proposed a denoising technique which is independent on sampling conversion, so that texture features can be maintained, in which the speckle noise is reduced to the maximum extent.

show abstract

Two-stage image denoising by principal component analysis with local pixel grouping

Cited by 559 publications

References 26 publications

MRI noise estimation and denoising using non-local PCA

MRI noise estimation and denoising using non-local PCA

A New Approach to Image Denoising by Patch-Based Algorithm

RISAT-1 Image Despeckling using a Modified Undecimated BM3D Technique

Contact Info

Product

Resources

About