Adaptive Denoising by Singular Value Decomposition
Authors:
- 28 January 2011
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 18 (4) , 215-218
Abstract
This letter presents an adaptive denoising method based on the singular value decomposition (SVD). By incorporating a global subspace analysis into the scheme of local basis selection, the problems of previous adaptive methods are effectively tackled. Experimental results show that the proposed method achieves outstanding preservation of image details, and at high noise levels it provides improvements in both objective and subjective quality of the denoised image when compared to the state-of-the-art methods.Keywords
This publication has 10 references indexed in Scilit:
- Two-stage image denoising by principal component analysis with local pixel groupingPattern Recognition, 2010
- Fast Non-Local Means (NLM) Computation With Probabilistic Early TerminationIEEE Signal Processing Letters, 2009
- Principal Neighborhood Dictionaries for Nonlocal Means Image DenoisingIEEE Transactions on Image Processing, 2009
- SURE-Based Non-Local MeansIEEE Signal Processing Letters, 2009
- Non-Parametric Detection of the Number of Signals: Hypothesis Testing and Random Matrix TheoryIEEE Transactions on Signal Processing, 2009
- Image Denoising by Sparse 3-D Transform-Domain Collaborative FilteringIEEE Transactions on Image Processing, 2007
- Image Denoising Via Sparse and Redundant Representations Over Learned DictionariesIEEE Transactions on Image Processing, 2006
- A Non-Local Algorithm for Image Denoising2005
- Adaptive principal components and image denoising2004
- Ideal Spatial Adaptation by Wavelet ShrinkageBiometrika, 1994