Hyperspectral Image Denoising With Group Sparse and Low-Rank Tensor Decomposition
| Autor: | Huali Li, Shutao Li, Jon Atli Benediktsson, Leyuan Fang, Zhihong Huang |
|---|---|
| Přispěvatelé: | Rafmagns- og tölvuverkfræðideild (HÍ), Faculty of Electrical and Computer Engineering (UI), Verkfræði- og náttúruvísindasvið (HÍ), School of Engineering and Natural Sciences (UI), Háskóli Íslands, University of Iceland |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
General Computer Science
Rank (linear algebra) Computer science Noise reduction 0211 other engineering and technologies 02 engineering and technology Iterative reconstruction Myndvinnsla Impulse noise Matrix decomposition Matrix (mathematics) symbols.namesake Hyperspectral image denoising 0202 electrical engineering electronic engineering information engineering General Materials Science sparse and low-rank tensor decomposition Tensor 021101 geological & geomatics engineering Sparse matrix Nonlocal similarity Denoising business.industry General Engineering Hyperspectral imaging Pattern recognition nonlocal similarity Gaussian noise Sparse and low-rank tensor decomposition symbols Physics::Accelerator Physics 020201 artificial intelligence & image processing lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence business lcsh:TK1-9971 |
| Zdroj: | IEEE Access, Vol 6, Pp 1380-1390 (2018) |
| Popis: | Hyperspectral image (HSI) is usually corrupted by various types of noise, including Gaussian noise, impulse noise, stripes, deadlines, and so on. Recently, sparse and low-rank matrix decomposition (SLRMD) has demonstrated to be an effective tool in HSI denoising. However, the matrix-based SLRMD technique cannot fully take the advantage of spatial and spectral information in a 3-D HSI data. In this paper, a novel group sparse and low-rank tensor decomposition (GSLRTD) method is proposed to remove different kinds of noise in HSI, while still well preserving spectral and spatial characteristics. Since a clean 3-D HSI data can be regarded as a 3-D tensor, the proposed GSLRTD method formulates a HSI recovery problem into a sparse and low-rank tensor decomposition framework. Specifically, the HSI is first divided into a set of overlapping 3-D tensor cubes, which are then clustered into groups by K-means algorithm. Then, each group contains similar tensor cubes, which can be constructed as a new tensor by unfolding these similar tensors into a set of matrices and stacking them. Finally, the SLRTD model is introduced to generate noisefree estimation for each group tensor. By aggregating all reconstructed group tensors, we can reconstruct a denoised HSI. Experiments on both simulated and real HSI data sets demonstrate the effectiveness of the proposed method. This paper was supported in part by the National Natural Science Foundation of China under Grant 61301255, Grant 61771192, and Grant 61471167, in part by the National Natural Science Fund of China for Distinguished Young Scholars under Grant 61325007, in part by the National Natural Science Fund of China for International Cooperation and Exchanges under Grant 61520106001, and in part by the Science and Technology Plan Project Fund of Hunan Province under Grant 2015WK3001 and Grant 2017RS3024. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|