Titill: | Hyperspectral Image Denoising With Group Sparse and Low-Rank Tensor Decomposition |
Höfundur: |
|
Útgáfa: | 2018 |
Tungumál: | Enska |
Umfang: | 1380-1390 |
Háskóli/Stofnun: | Háskóli Íslands University of Iceland |
Svið: | Verkfræði- og náttúruvísindasvið (HÍ) School of Engineering and Natural Sciences (UI) |
Deild: | Rafmagns- og tölvuverkfræðideild (HÍ) Faculty of Electrical and Computer Engineering (UI) |
Birtist í: | IEEE Access;6 |
ISSN: | 2169-3536 |
DOI: | 10.1109/ACCESS.2017.2778947 |
Efnisorð: | Hyperspectral image; Denoising; Sparse and low-rank tensor decomposition; Nonlocal similarity; Myndvinnsla |
URI: | https://hdl.handle.net/20.500.11815/821 |
Tilvitnun:Huang, Z., Li, S., Fang, L., Li, H., & Benediktsson, J. A. (2018). Hyperspectral Image Denoising With Group Sparse and Low-Rank Tensor Decomposition. IEEE Access, 6, 1380-1390. doi:10.1109/ACCESS.2017.2778947
|
|
Útdráttur: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.
|