浏览全部资源
扫码关注微信
Received：11 January 2021，
Revised：2021-4-2，
Accepted：09 April 2021，
Published：16 August 2021
移动端阅览
目的
2
高光谱图像复原是高光谱领域中一个重要的预处理步骤，能够有效去除成像条件所带来的不利影响，提升后续处理任务的精度。张量核范数被广泛应用于高光谱复原问题中，得到了较好的结果。然而，在张量核范数的定义中，它对张量所有奇异值使用相同的阈值进行收缩，未充分考虑高光谱的物理意义，得到了次优的结果。为了提升高光谱图像复原的精度，本文提出了基于频率加权张量核范数的高光谱复原算法。
方法
2
在张量的频率域内，对清晰的高光谱图像添加噪声，图像信息在低频部分变化较小，而在高频部分变化巨大。基于这样的物理意义，定义了一种频率加权张量核范数来逼近张量秩函数，提出了频率域权重的自适应确定方法，让其能减少对低频部分的收缩，同时加大高频部分惩罚。然后将其应用于高光谱图像复原和去噪问题中，并基于交替方向乘子法设计了相应最小化问题的快速求解算法。
结果
2
在4个高光谱数据集上与相关方法进行对比仿真实验，高采样率条件下在Washington DC Mall数据集上，相比性能第2的模型，本文模型复原结果的PSNR（peak signal-to-noise ratio）提升了1.76 dB；在Stuff数据集上，PSNR值提升了2.91 dB。高噪声条件下，在Pavia数据集上相比性能第2的模型，本文模型去噪结果的PSNR提升了8.61 dB；在Indian数据集上，PSNR值提升了10.77 dB。
结论
2
本文模型可以更好地探索高光谱图像的低秩特性，使复原的图像在保持主体信息的同时，复原出更多图像纹理细节。
Objective
2
Hyperspectral images (HSIs) use imaging spectrometers to collect hundreds of spectral band images from ultraviolet to infrared wavelengths on the same area of the earth's surface. Tens to hundreds of continuous grayscale images are available
and each pixel can extract a spectral curve. Hyperspectral imaging technology closely combines the traditional 2D image remote sensing technology and spectroscopy technology. While acquiring the spatial information of the ground object
it arranges the radiation energy of the measured object according to the wavelength and obtains hundreds of continuous data in a spectral range. It is widely used in various fields
such as environmental monitoring and terrain classification. However
real HSI is often subject to different types of pollution
namely
noise
undersampling
or missing data
because of actual imaging conditions
weather conditions
or data transmission procedures. These types of pollution severely reduce the quality of HSI and limit the accuracy of subsequent processing tasks
such as unmixing and target detection. Restoration from a noisy
undersampling
or incomplete HSI is an ill-posed inverse problem. A common method is to treat each band as a grayscale image or column HSI into 2D matrix and use the method of matrix restoration to process it. Rank minimization is a common strategy to solve such problems. However
the rank function is discrete and non-convex
and the solution of the rank function minimization problem is an nondeterministic polynominal(NP)-hard problem. The convex envelope-nuclear norm of the rank function is usually used to approximate the rank function for solving this problem
and the rank minimization problem is transformed into the nuclear norm minimization problem. Although nuclear norm can achieve good restoration results
it imposes the same shrinkage on all singular values of the matrix due to the limitation of convexity. This imposition results in a finite degree approximation of the rank function. At the same time
HSI is an imaging of the same scene under different spectra
and its data in each band are highly correlated. This correlation is often ignored. The exploration of the low-rank characteristics of HSI by an approach of the column matrix is insufficient.
Method
2
One of the methods to define the rank of the tensor is tensor singular value decomposition
which can also be solved by the tensor nuclear norm. However
all singular values are treated equally in the definition of the tensor nuclear norm
and the frequency information of the tensor is ignored. In the frequency domain of the tensor
noises are added to the clear hyperspectral image. The image information changes slowly in the low-frequency forward slicing matrix
but it changes markedly in the high-frequency forward slicing matrix. In this study
we propose a frequency-weighted tensor nuclear norm based on the physical properties of tensors in the frequency domain. The original tensor is protected under the condition of removing outliers by appropriately reducing the penalty for the low-frequency forward slicing matrix. The main information is to increase the penalty for the high-frequency forward slicing matrix
which can fully remove the outliers in the tensor. Ultimately
the degree of approximation of the tensor nuclear norm to the tensor rank is improved to enhance the accuracy of restoration. At the same time
we explore the changes in the frequency components of the hyperspectral image when the sparse noise is disturbed through numerical simulation experiments. The sparse noise slightly affects the nuclear norm of the low-frequency band slice matrix
but it influences the high-frequency band slice matrix nuclear norm. The nuclear norm of the low-frequency band slice matrix changes slowly when the noise intensity increases
while that of the high-frequency band slice matrix changes markedly. On this basis
we give the weight adaptive calculation method of each frequency forward slice matrix
increase the adaptive data based on the frequency-weighted tensor nuclear norm
and reduce the human intervention in the parameter adjustment process to ensure robustness of the model.
Result
2
We compare simulation experiments with related methods on four hyperspectral datasets. On the simulated dataset
the sampling rates are 10%
20%
and 30%. Compared with the indices of the second best performing model
the peak signal-to-noise ratio (PSNR) index and the structure similarity (SSIM) index on the Washington DC Mall dataset can be increased by 0.98 dB
1.64 dB
and 1.76 dB
and 0.005 8
0.019 3
and 0.010 2. Compared with the indices of the second best performing model
the PSNR and SSIM values of the restoration result of the proposed model on the Stuff dataset can be increased by 1.84 dB
2.65 dB
2.91 dB
and 0.085 2
0.042 5
and 0.023 1. The sparse noise intensity values are 0.5 and 0.4. Compared with the indices of the second best performing model
the PSNR and SSIM values of the denoising results of the proposed model on the Pavia dataset can be increased by 8.61 dB and 6.67 dB
and 0.441 and 0.087 41. Compared with the indices of the second best performing model
the PSNR and SSIM values of the denoising results of the proposed model on the Indian dataset can be increased by 10.77 dB and 6.34 dB and 0.403 and 0.033 1.
Conclusion
2
The proposed model considers the original HSI information carried by different frequency slices and the influence of missing or sparse noise on the nuclear norm of each frequency slice. It can better explore the low-rank characteristics of hyperspectral images. Thus
the restored HSI can be maintained. More texture details are restored simultaneously with the main information.
Boyd S, Parikh N, Chu E, Peleato B and Eckstein J. 2011. Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundations and Trends® in Machine Learning, 3(1): 1-122[DOI: 10.1561/2200000016]
Cai J F, Candès E J and Shen Z W. 2010. A singular value thresholding algorithm for matrix completion. SIAM Journal on Optimization, 20(4): 1956-1982[DOI: 10.1137/080738970]
Fazel M. 2002. Matrix Rank Minimization with Applications. Stanford University, USA
Frick A and Tervooren S. 2019. A framework for the long-term monitoring of urban green volume based on multi-temporal and multi-sensoral remote sensing data. Journal of Geovisualization and Spatial Analysis, 3(1): #6[DOI: 10.1007/s41651-019-0030-5]
Friedland S and Lim L H. 2018. Nuclear norm of higher-order tensors. Mathematics of Computation, 87: 1255-1281[DOI: 10.1090/mcom/3239]
Gu S H, Xie Q, Meng D Y, Zuo W M, Feng X C and Zhang L. 2017. Weighted nuclear norm minimization and its applications to low level vision. International Journal of Computer Vision, 121(2): 183-208[DOI: 10.1007/s11263-016-0930-5]
He W, Zhang H Y, Shen H F and Zhang L P. 2018. Hyperspectral image denoising using local low-rank matrix recovery and global spatial-spectral total variation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(3): 713-729[DOI: 10.1109/JSTARS.2018.2800701]
He W, Zhang H Y, Zhang L P and Shen H F. 2016. Total-variation-regularized low-rank matrix factorization for hyperspectral image restoration. IEEE Transactions on Geoscience and Remote Sensing, 54(1): 178-188[DOI: 10.1109/TGRS.2015.2452812]
Hitchcock F L. 1927. The expression of a tensor or a polyadic as a sum of products. Journal of Mathematics and Physics, 6(1/4): 164-189[DOI: 10.1002/sapm192761164]
Hong D F, Gao L R, Yao J, Zhang B, Plaza A and Chanussot J. 2021a. Graph convolutional networks for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 59(7): 5966-5978[DOI: 10.1109/TGRS.2020.3015157]
Hong D F, Gao L R, Yokoya N, Yao J, Chanussot J, Du Q and Zhang B. 2021b. More diverse means better: multimodal deep learning meets remote-sensing imagery classification. IEEE Transactions on Geoscience and Remote Sensing, 59(5): 4340-4354[DOI: 10.1109/TGRS.2020.3016820]
Hong D F, Yokoya N, Chanussot J and Zhu X X. 2019a. An augmented linear mixing model to address spectral variability for hyperspectral unmixing. IEEE Transactions on Image Processing, 28(4): 1923-1938[DOI: 10.1109/TIP.2018.2878958]
Hong DF, Yokoya N, Ge N, Chanussot J and Zhu X X. 2019b. Learnable manifold alignment (LeMA): a semi-supervised cross-modality learning framework for land cover and land use classification. ISPRS Journal of Photogrammetry and Remote Sensing, 147: 193-205[DOI: 10.1016/j.isprsjprs.2018.10.006]
Ji S W and Ye J P. 2009. An accelerated gradient method for trace norm minimization//Proceedings of the 26th Annual International Conference on Machine Learning. Montreal, Canada: ACM: 457-464[ DOI: 10.1145/1553374.1553434 http://dx.doi.org/10.1145/1553374.1553434 ]
Ji T Y, Huang T Z, Zhao X L and Sun D L. 2017. A new surrogate for tensor multirank and applications in image and video completion//Proceedings of 2017 International Conference on Progress in Informatics and Computing (PIC). Nanjing, China: IEEE: 101-107[ DOI: 10.1109/PIC.2017.8359523 http://dx.doi.org/10.1109/PIC.2017.8359523 ]
Jiang T X, Huang T Z, Zhao X L and Deng L J. 2020. Multi-dimensionalimaging data recovery via minimizing the partial sum of tubal nuclear norm. Journal of Computational and Applied Mathematics, 372: #112680[DOI: 10.1016/j.cam.2019.112680]
Kilmer M E, Braman K, Hao N and Hoover R C. 2013. Third-order tensors as operators on matrices: a theoretical and computational framework with applications in imaging. SIAM Journal on Matrix Analysis and Applications, 34(1): 148-172[DOI: 10.1137/110837711]
Li Z K, Wang T N, Li W, Du Q, Wang C Y, Liu C W and Shi X B. 2020. Deep multilayer fusion dense network for hyperspectral image classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13: 1258-1270[DOI: 10.1109/JSTARS.2020.2982614]
Lin C, Chen S Y, Chen C C and Tai C H. 2018. Detecting newly grown tree leaves from unmanned-aerial-vehicle images using hyperspectral target detection techniques. ISPRS Journal of Photogrammetry and Remote Sensing, 142: 174-189[DOI: 10.1016/j.isprsjprs.2018.05.022]
Lin Z C, Ganesh A, Wright J, Wu L Q, Chen M M and Ma Y. 2009. Fast convex optimization algorithms for exact recovery of a corrupted low-rank matrix. Coordinated Science Laboratory Report No. UILU-ENG-09-2214, DC-246. Coordinated Science Laboratory, University of Illinois at Urbana-Champaign
Liu J, Musialski P, Wonka P and Ye J P. 2013. Tensor completion for estimating missing values in visual data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(1): 208-220[DOI: 10.1109/TPAMI.2012.39]
Liu N, Li L, Li W, Tao R, Fowler J E and Chanussot J. 2021. Hyperspectral restoration and fusion with multispectral imagery via low-rank tensor-approximation. IEEE Transactions on Geoscience and Remote Sensing[DOI: 10.1109/TGRS.2020.3049014]
Liu N, Li W, Tao R and Fowler J E. 2019. Wavelet-domain low-rank/group-sparse destriping for hyperspectral imagery. IEEE Transactions on Geoscience and Remote Sensing, 57(12): 10310-10321[DOI: 10.1109/TGRS.2019.2933555]
Lu C Y, Feng J S, Chen Y D, Liu W, Lin Z C and Yan S C. 2020. Tensor robust principal component analysis with a new tensor nuclear norm. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(4): 925-938[DOI: 10.1109/TPAMI.2019.2891760]
Ran Q, Yu H Y, GaoL R, Li W and Zhang B. 2018. Superpixel and subspace projection-based support vector machines for hyperspectral image classification. Journal of Image and Graphics, 23(1): 95-105
冉琼, 于浩洋, 高连如, 李伟, 张兵. 2018. 结合超像元和子空间投影支持向量机的高光谱图像分类. 中国图象图形学报, 23(1): 95-105 [DOI: 10.11834/jig.170201]
Semerci O, Hao N, Kilmer M E and Miller E L. 2014. Tensor-based formulation and nuclear norm regularization for multienergy computed tomography. IEEE Transactions on Image Processing, 23(4): 1678-1693[DOI: 10.1109/TIP.2014.2305840]
Tan K, Wang X and Du P J. 2019. Research progress of the remote sensing classification combining deep learning and semi-supervised learning. Journal of Image and Graphics, 24(11): 1823-1841
谭琨, 王雪, 杜培军. 2019. 结合深度学习和半监督学习的遥感影像分类进展. 中国图象图形学报, 24(11): 1823-1841 [DOI: 10.11834/jig.190348]
Tucker L R. 1966. Some mathematical notes on three-mode factor analysis. Psychometrika, 31(3): 279-311[DOI: 10.1007/BF02289464]
Wang S H, Liu Y P, Feng L L and Zhu C. 2020. Frequency-weighted robust tensor principal component analysis[EB/OL]. https://arxiv.org/pdf/2004.10068.pdf https://arxiv.org/pdf/2004.10068.pdf
Wang Z, Bovik A C, Sheikh H R and Simoncelli E P. 2004. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4): 600-612[DOI: 10.1109/TIP.2003.819861]
Xie Q, Zhao Q, Meng D Y, Xu Z B, Gu S H, Zuo W M and Zhang L. 2016. Multispectral images denoising by intrinsic tensor sparsity regularization//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE: 1692-1700[ DOI: 10.1109/CVPR.2016.187 http://dx.doi.org/10.1109/CVPR.2016.187 ]
Xie T, Li S T, Fang L Y and Liu L C. 2019. Tensor completion via nonlocal low-rank regularization. IEEE Transactions on Cybernetics, 49(6): 2344-2354[DOI: 10.1109/TCYB.2018.2825598]
Yang R Y, Jia Y X, Xu P and Xie X Z. 2019. Hyperspectral image restoration with truncated nuclear norm minimization and total variation regularization. Journal of Image and Graphics, 24(10): 1801-1812
杨润宇, 贾亦雄, 徐鹏, 谢晓振. 2019. 截断核范数和全变差正则化高光谱图像复原. 中国图象图形学报, 24(10): 1801-1812 [DOI: 10.11834/jig.180433]
Zeng H J, Chen Y Y, Xie X Z and Ning J F. 2021b. Enhanced nonconvex low-rank approximation of tensor multi-modes for tensor completion. IEEE Transactions on Computational Imaging, 7: 164-177[DOI: 10.1109/TCI.2021.3053699]
Zeng H J, Xie X Z, Cui H J, Zhao Y and Ning J F. 2020. Hyperspectral image restoration via CNN denoiser prior regularized low-rank tensor recovery. ComputerVision and Image Understanding, 197: #103004[DOI: 10.1016/j.cviu.2020.103004]
Zeng H J, Xie X Z and Ning J F. 2021a. Hyperspectral image denoising via global spatial-spectral total variation regularized nonconvex local low-rank tensor approximation. Signal Processing, 178: #107805[DOI: 10.1016/j.sigpro.2020.107805]
Zhang C H. 2010. Nearly unbiased variable selection under minimax concave penalty. The Annals of Statistics, 38(2): 894-942[DOI: 10.1214/09-AOS729]
Zhang H Y, Liu L, He W and Zhang L P. 2020a. Hyperspectral image denoising with total variation regularization and nonlocal low-rank tensor decomposition. IEEE Transactions on Geoscience and Remote Sensing, 58(5): 3071-3084[DOI: 10.1109/TGRS.2019.2947333]
Zhang Y, Fan Y G, Xu M M, Li W, Zhang G Y, Liu L and Yu D F. 2020b. An improved low rank and sparse matrix decomposition-based anomaly target detection algorithm for hyperspectral imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13: 2663-2672[DOI: 10.1109/JSTARS.2020.2994340]
Zhou P, Lu C Y, Lin Z C and Zhang C. 2018. Tensor factorization for low-rank tensor completion. IEEE Transactions on Image Processing, 27(3): 1152-1163[DOI:10.1109/TIP.2017.2762595]
相关文章
相关作者
相关机构
京公网安备11010802024621
