Adaptive Rank and Structured Sparsity Corrections for Hyperspectral Image Restoration
- 19 February 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 52 (9) , 8729-8740
Abstract
Hyperspectral images (HSIs) are inevitably contaminated by the mixed noise (such as Gaussian noise, impulse noise, deadlines, and stripes), which could influence the subsequent processing accuracy. Generally, HSI restoration can be transformed into the low-rank matrix recovery (LRMR). In the LRMR, the nuclear norm is widely used to substitute the matrix rank, but its effectiveness is still worth improving. Besides, the $l_{0}$ -norm cannot capture the sparse noise’s structured sparsity property. To handle these issues, the adaptive rank and structured sparsity corrections (ARSSC) are presented for HSI restoration. The ARSSC introduces two convex regularizers, that is: 1) the rank correction (RC) and 2) the structured sparsity correction (SSC), to, respectively, approximate the matrix rank and the $l_{2,0}$ -norm. The RC and the SSC can adaptively offset the penalization of large entries from the nuclear norm and the $l_{2,1}$ -norm, respectively, where the larger the entry, the greater its offset. Therefore, the proposed ARSSC achieves a tighter approximation of the noise-free HSI low-rank structure and promotes the structured sparsity of sparse noise. An efficient alternative direction method of multipliers (ADMM) algorithm is applied to solve the resulting convex optimization problem. The superiority of the ARSSC in terms of the mixed noise removal and spatial-spectral structure information preserving, is demonstrated by several experimental results both on simulated and real datasets, compared with other state-of-the-art HSI restoration approaches.Keywords
Funding Information
- Major Program of the National Natural Science Fund of China (61890962)
- National Natural Science Fund of China for International Cooperation and Exchanges (61520106001)
- Fund of Hunan Province for Science and Technology Plan Project (2017RS3024)
- Fund of Key Laboratory of Visual Perception and Artificial Intelligence of Hunan Province (2018TP1013)
- Science and Technology Talents Program of Hunan Association for Science and Technology (2017TJ-Q09)
This publication has 48 references indexed in Scilit:
- Hyperspectral Target Detection : An Overview of Current and Future ChallengesIEEE Signal Processing Magazine, 2013
- Most Tensor Problems Are NP-HardJournal of the ACM, 2013
- Hyperspectral Image Restoration Using Low-Rank Matrix RecoveryIEEE Transactions on Geoscience and Remote Sensing, 2013
- Integration of Full and Mixed Pixel Techniques to Obtain Thematic Maps with a Re†ned Resolution2012
- Robust principal component analysis?Journal of the ACM, 2011
- An introduction to hyperspectral imaging and its application for security, surveillance and target acquisitionThe Imaging Science Journal, 2010
- A Singular Value Thresholding Algorithm for Matrix CompletionSIAM Journal on Optimization, 2010
- Hyperspectral image data analysisIEEE Signal Processing Magazine, 2002
- Characterization of the subdifferential of some matrix normsLinear Algebra and its Applications, 1992
- On the Douglas—Rachford splitting method and the proximal point algorithm for maximal monotone operatorsMathematical Programming, 1992