Hyperspectral Image Denoising Based on Nonlocal Low-Rank and TV Regularization

Xiangyang Kong,Haixia Huang,Jize Xue,Jonathan Cheung-Wai Chan,Jiyuan Zang,Zhigang Ren,Yongqiang Zhao

doi:10.3390/rs12121956

Abstract

Hyperspectral image (HSI) acquisitions are degraded by various noises, among which additive Gaussian noise may be the worst-case, as suggested by information theory. In this paper, we present a novel tensor-based HSI denoising approach by fully identifying the intrinsic structures of the clean HSI and the noise. Specifically, the HSI is first divided into local overlapping full-band patches (FBPs), then the nonlocal similar patches in each group are unfolded and stacked into a new third order tensor. As this tensor shows a stronger low-rank property than the original degraded HSI, the tensor weighted nuclear norm minimization (TWNNM) on the constructed tensor can effectively separate the low-rank clean HSI patches. In addition, a regularization strategy with spatial–spectral total variation (SSTV) is utilized to ensure the global spatial–spectral smoothness in both spatial and spectral domains. Our method is designed to model the spatial–spectral non-local self-similarity and global spatial–spectral smoothness simultaneously. Experiments conducted on simulated and real datasets show the superiority of the proposed method.

Highlights

For various hyperspectral image (HSI) applications, it is important to fully exploit useful spatial–spectral features of HSI
For the weight parameters λi (i = 1, 2, 3) in total variation (TV) regularization, considering that λ1 and λ2 both control the spatial dimension of HSI, they should be assigned to the same weights
FigFuigreur1e1.11V.e(V1r)teiPrctaairclaamml emeteaernasnspeprleorcfiotifloielnsesooffththeeddeennooiissiinngg results ooff tthheesseeccoonnddssppecetcrtarlabl abnadndininthtehAe iArbiorbrnoerne VisVibisleib/Ilne/fIrnafrTreahdreeIrdemgIuamlgaiarnigzgiantSigopneSpcpateroacmtmroeetmeteresrtae(rAreVd(AiIsRcVuIISsRs)eISdSa)inliSnSaealcistniioamns 4ai.mg1.eaT:gh(ea:)ot(ohare)irgotiwrniogalipnabaraalnmbdea,tne(rdbs,)arH(ebyr)eRlHaetyse,dR(ecs),L(Rc)TV, (d)LLRRTMV,tRo(,dt(h)eeL)pRNaMtAchRIL,(s(RiezM)eNmAA,×I(Lmf)R×LMRbA)T,Dth(feT)VpLa,RraaTnmDdeTt(egVrs), ipannrvodoplv(ogesd)epadr.eopmoasend.k, and k is the number of similar patches in each cluster

Summary

Introduction

For various hyperspectral image (HSI) applications, it is important to fully exploit useful spatial–spectral features of HSI. Unfolding all of the bands in HSI to a long vector is done by the vector method [7] This kind of method has a high processing speed at the cost of destroying the spatial structure and spectral correlation. The matrix method can be divided into the following two categories: band-by-band method and tensor-matrixing method [8] The former is a natural generalization of the procession of a gray-level image. Tensor-matrixing is conducted by unfolding each band into a vector, and all of the vectors are cascaded into a matrix This kind of method considers spectral correlation, the spatial structure could still be destroyed. Some advanced techniques in traditional image processing have been adopted for HSI denoising, such as nonlocal similarity [11] and anisotropic diffusion [12]

Results

Discussion

Conclusion