Nonlocal low-rank regularized tensor decomposition for hyperspectral image denoising
- Author
- Jize Xue, Yongqiang Zhao, Wenzhi Liao (UGent) and Jonathan Cheung-Wai Chan
- Organization
- Abstract
- Hyperspectral image (HSI) enjoys great advantages over more traditional image types for various applications due to the extra knowledge available. For the nonideal optical and electronic devices, HSI is always corrupted by various noises, such as Gaussian noise, deadlines, and stripings. The global correlation across spectrum (GCS) and nonlocal self-similarity (NSS) over space are two important characteristics for HSI. In this paper, a nonlocal low-rank regularized CANDECOMP/PARAFAC (CP) tensor decomposition (NLR-CPTD) is proposed to fully utilize these two intrinsic priors. To make the rank estimation more accurate, a new manner of rank determination for the NLR-CPTD model is proposed. The intrinsic GCS and NSS priors can be efficiently explored under the low-rank regularized CPTD to avoid tensor rank estimation bias for denoising performance. Then, the proposed HSI denoising model is performed on tensors formed by nonlocal similar patches within an HSI. The alternating direction method of multipliers-based optimization technique is designed to solve the minimum problem. Compared with state-of-the-art methods, the proposed algorithm can greatly promote the denoising performance of an HSI in various quality assessments.
- Keywords
- Electrical and Electronic Engineering, General Earth and Planetary Sciences, CANDECOMP/PARAFAC (CP) tensor decomposition (CPTD), hyperspectral image (HSI) denoising, nonlocal low-rank regularization (LR), rank automatic determination, rank estimation bias, SPARSE REPRESENTATION, NOISE-REDUCTION, RESTORATION, MODEL
Downloads
-
(...).pdf
- full text (Published version)
- |
- UGent only
- |
- |
- 11.46 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-8667047
- MLA
- Xue, Jize, et al. “Nonlocal Low-Rank Regularized Tensor Decomposition for Hyperspectral Image Denoising.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 57, no. 7, 2019, pp. 5174–89, doi:10.1109/tgrs.2019.2897316.
- APA
- Xue, J., Zhao, Y., Liao, W., & Chan, J. C.-W. (2019). Nonlocal low-rank regularized tensor decomposition for hyperspectral image denoising. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 57(7), 5174–5189. https://doi.org/10.1109/tgrs.2019.2897316
- Chicago author-date
- Xue, Jize, Yongqiang Zhao, Wenzhi Liao, and Jonathan Cheung-Wai Chan. 2019. “Nonlocal Low-Rank Regularized Tensor Decomposition for Hyperspectral Image Denoising.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 57 (7): 5174–89. https://doi.org/10.1109/tgrs.2019.2897316.
- Chicago author-date (all authors)
- Xue, Jize, Yongqiang Zhao, Wenzhi Liao, and Jonathan Cheung-Wai Chan. 2019. “Nonlocal Low-Rank Regularized Tensor Decomposition for Hyperspectral Image Denoising.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 57 (7): 5174–5189. doi:10.1109/tgrs.2019.2897316.
- Vancouver
- 1.Xue J, Zhao Y, Liao W, Chan JC-W. Nonlocal low-rank regularized tensor decomposition for hyperspectral image denoising. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING. 2019;57(7):5174–89.
- IEEE
- [1]J. Xue, Y. Zhao, W. Liao, and J. C.-W. Chan, “Nonlocal low-rank regularized tensor decomposition for hyperspectral image denoising,” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 57, no. 7, pp. 5174–5189, 2019.
@article{8667047,
abstract = {{Hyperspectral image (HSI) enjoys great advantages over more traditional image types for various applications due to the extra knowledge available. For the nonideal optical and electronic devices, HSI is always corrupted by various noises, such as Gaussian noise, deadlines, and stripings. The global correlation across spectrum (GCS) and nonlocal self-similarity (NSS) over space are two important characteristics for HSI. In this paper, a nonlocal low-rank regularized CANDECOMP/PARAFAC (CP) tensor decomposition (NLR-CPTD) is proposed to fully utilize these two intrinsic priors. To make the rank estimation more accurate, a new manner of rank determination for the NLR-CPTD model is proposed. The intrinsic GCS and NSS priors can be efficiently explored under the low-rank regularized CPTD to avoid tensor rank estimation bias for denoising performance. Then, the proposed HSI denoising model is performed on tensors formed by nonlocal similar patches within an HSI. The alternating direction method of multipliers-based optimization technique is designed to solve the minimum problem. Compared with state-of-the-art methods, the proposed algorithm can greatly promote the denoising performance of an HSI in various quality assessments.}},
author = {{Xue, Jize and Zhao, Yongqiang and Liao, Wenzhi and Chan, Jonathan Cheung-Wai}},
issn = {{0196-2892}},
journal = {{IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING}},
keywords = {{Electrical and Electronic Engineering,General Earth and Planetary Sciences,CANDECOMP/PARAFAC (CP) tensor decomposition (CPTD),hyperspectral image (HSI) denoising,nonlocal low-rank regularization (LR),rank automatic determination,rank estimation bias,SPARSE REPRESENTATION,NOISE-REDUCTION,RESTORATION,MODEL}},
language = {{eng}},
number = {{7}},
pages = {{5174--5189}},
title = {{Nonlocal low-rank regularized tensor decomposition for hyperspectral image denoising}},
url = {{http://doi.org/10.1109/tgrs.2019.2897316}},
volume = {{57}},
year = {{2019}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: