Typical Application of Graph Signal Processing in Hyperspectral Image Processing
-
摘要: 高光谱图像(HSI)具有纳米级的光谱分辨能力且同时对地物目标的光谱维和空间维进行联合成像的优势，能够精细化感知场景目标的本征判别属性，在遥感探测、医疗诊断和国防安全等具有重要应用价值，是高精度遥感探测的科技制高点之一。不同于传统1维时间信号、2维图像信号，高光谱图像具有多阶、高维的信号属性。为解决传统信号处理方法在高光谱图像处理领域中的不足，图信号处理(GSP)理论与方法被逐渐引入高光谱图像处理与解译等任务中。该文以短综述的形式，介绍了图信号处理在高光谱图像处理领域的理论发展并列举了在高光谱特征提取、图像重构和解译分类3个主要方面的典型应用。最后，进一步探讨了该方向未来发展所面临的挑战和相应解决办法。Abstract: HyperSpectral Image(HSI) has nanometer-level spectral discriminative ability, capturing the spectral and spatial information of the ground objects simultaneously, within the integration of three-dimensional image cube. The capability to finely sense the intrinsic properties of objects makes it universally applied to many fields, e.g., remote sensing & detection, medical imaging & diagnosis, military defense & security, etc. Different from traditional one-dimensional time-series signals and two-dimensional image signals, HSIs are third-order tensor signals, with the spectral bands in the third-mode being high-dimensional. To eliminate the deficiencies of existing techniques in solving HSI processing and interpretation problems, Graph Signal Processing (GSP) is introduced. A short overview of the theoretical and technological development of GSP is given, along with its typical applications in HSI feature extraction, restoration, and classification. Based on the survey of the existing research basis, the future challenges and potential approaches to solve them in the community are also pointed out and discussed.
-
算法1 基于SLRG的有监督高光谱特征提取算法 (1) 输入${\boldsymbol{X}}$，特征提取维数$ k $ (2) for $ i = 1:C $ do (3) for $ j{\text{ = 1}}:{c_i} $($ {c_i} $表示第$ i $类样本个数) do (4) 根据式(9)对每个${\boldsymbol{x}}_j^{(i)}$用其同一类别的训练数据求稀疏
低秩表示系数${\boldsymbol{w} }_j^{(i)}$(5) 同一类别的稀疏低秩表示矩阵${{\boldsymbol{W}}^{(i)} } = [{{\boldsymbol{W}}^{(i)} };{\boldsymbol{w}}_j^{(i)}]$ (6) end for (7) 构建稀疏低秩表示图${\boldsymbol{W} } = {\text{diag} }({ {\boldsymbol{W} }^{(1)} },{ {\boldsymbol{W} }^{(2)} }, \cdots ,{ {\boldsymbol{W} }^{(C)} })$ (8) end for (9) 根据式(10)求得${\boldsymbol{P}}$ (10) 输出${\boldsymbol{Y}} = {{\boldsymbol{P}}^{\text{T} } }{\boldsymbol{X}}$ 
下载: 导出CSV
-
[1] GOETZ A F H, VANE G, SOLOMON J E, et al. Imaging spectrometry for earth remote sensing[J]. Science, 1985, 228(4704): 1147–1153. doi: 10.1126/science.228.4704.1147 [2] BIOUCAS-DIAS J M, PLAZA A, CAMPS-VALLS G, et al. Hyperspectral remote sensing data analysis and future challenges[J]. IEEE Geoscience and Remote Sensing Magazine, 2013, 1(2): 6–36. doi: 10.1109/MGRS.2013.2244672 [3] 王茂芝. 高光谱遥感影像处理与地质应用若干关键问题研究[D]. [博士论文], 成都理工大学, 2014.WANG Maozhi. Researches on several critical problems of hyperspectral remote sensing image processing and geologic application[D]. [Ph. D. dissertation], Chengdu University of Technology, 2014. [4] 李畅. 高光谱遥感影像处理中的若干关键技术研究[D]. [博士论文], 华中科技大学, 2018.LI Chang. Research on key technologies of hyperspectral remote sensing imagery[D]. [Ph. D. dissertation], Huazhong University of Science and Technology, 2018. [5] LIU Na, LI Wei, and DU Qian. Unsupervised feature extraction for hyperspectral imagery using collaboration-competition graph[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(6): 1491–1503. doi: 10.1109/JSTSP.2018.2877474 [6] LI Wei, FENG Fubiao, LI Hengchao, et al. Discriminant analysis-based dimension reduction for hyperspectral image classification: A survey of the most recent advances and an experimental comparison of different techniques[J]. IEEE Geoscience and Remote Sensing Magazine, 2018, 6(1): 15–34. doi: 10.1109/MGRS.2018.2793873 [7] RASTI B, HONG Danfeng, HANG Renlong, et al. Feature extraction for hyperspectral imagery: The evolution from shallow to deep: Overview and toolbox[J]. IEEE Geoscience and Remote Sensing Magazine, 2020, 8(4): 60–88. doi: 10.1109/MGRS.2020.2979764 [8] LIU Na, LI Wei, WANG Yinjian, et al. A survey on hyperspectral image restoration: From the view of low-rank tensor approximation[J]. arXiv: 2205.08839, 2022. [9] LIU Na, LI Wei, TAO Ran, et al. Wavelet-domain low-rank/group-sparse destriping for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(12): 10310–10321. doi: 10.1109/TGRS.2019.2933555 [10] GRIFFIN M K and BURKE H H K. Compensation of hyperspectral data for atmospheric effects[J]. Lincoln Laboratory Journal, 2003, 14(1): 29–54. [11] ZARE A and HO K C. Endmember variability in hyperspectral analysis: Addressing spectral variability during spectral unmixing[J]. IEEE Signal Processing Magazine, 2014, 31(1): 95–104. doi: 10.1109/MSP.2013.2279177 [12] CHANG C I. An information-theoretic approach to spectral variability, similarity, and discrimination for hyperspectral image analysis[J]. IEEE Transactions on Information Theory, 2000, 46(5): 1927–1932. doi: 10.1109/18.857802 [13] 薛朝辉. 高光谱遥感影像稀疏图嵌入分类研究[D]. [博士论文], 南京大学, 2015.XUE Zhaohui. Hyperspectral remote sensing image classification via sparse graph embedding[D]. [Ph. D. dissertation], Nanjing University, 2015. [14] SHUMAN D I, NARANG S K, FROSSARD P, et al. The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains[J]. IEEE Signal Processing Magazine, 2013, 30(3): 83–98. doi: 10.1109/MSP.2012.2235192 [15] DONG Xiaowen, THANOU D, TONI L, et al. Graph signal processing for machine learning: A review and new perspectives[J]. IEEE Signal Processing Magazine, 2020, 37(6): 117–127. doi: 10.1109/MSP.2020.3014591 [16] SELLARS P, AVILES-RIVERO A I, and SCHÖNLIEB C B. Superpixel contracted graph-based learning for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(6): 4180–4193. doi: 10.1109/TGRS.2019.2961599 [17] LUO Fulin, ZHANG Liangpei, DU Bo, et al. Dimensionality reduction with enhanced hybrid-graph discriminant learning for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(8): 5336–5353. doi: 10.1109/TGRS.2020.2963848 [18] 蒋俊正, 蔡万源. 一种基于图信号的高光谱图像去噪方法[P]. 中国专利, 202110216084.2, 2021.JIANG Junzheng and CAI Wanyuan. Graph-signal-based denoising for hyperspectral image[P]. China Patent, 202110216084.2, 2021. [19] WANG Si, HUANG Tingzhu, ZHAO Xile, et al. Double reweighted sparse regression and graph regularization for hyperspectral unmixing[J]. Remote Sensing, 2018, 10(7): 1046. doi: 10.3390/rs10071046 [20] BAI Jun, XIANG Shiming, and PAN Chunhong. A graph-based classification method for hyperspectral images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2013, 51(2): 803–817. doi: 10.1109/TGRS.2012.2205002 [21] ZHANG Si, TONG Hanghang, XU Jiejun, et al. Graph convolutional networks: A comprehensive review[J]. Computational Social Networks, 2019, 6(1): 11. doi: 10.1186/s40649-019-0069-y [22] SHAHRAKI F F and PRASAD S. Graph convolutional neural networks for hyperspectral data classification[C]. 2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP), Anaheim, USA, 2018: 968–972. [23] BATTAGLIA P W, HAMRICK J B, BAPST V, et al. Relational inductive biases, deep learning, and graph networks[J]. arXiv: 1806.01261, 2018. [24] XIA Feng, SUN Ke, YU Shuo, et al. Graph learning: A survey[J]. IEEE Transactions on Artificial Intelligence, 2021, 2(2): 109–127. doi: 10.1109/TAI.2021.3076021 [25] EGILMEZ H E, PAVEZ E, and ORTEGA A. Graph learning from data under Laplacian and structural constraints[J]. IEEE Journal of Selected Topics in Signal Processing, 2017, 11(6): 825–841. doi: 10.1109/JSTSP.2017.2726975 [26] 王保云, 李沛. 分析大数据: 非规则结构与图信号[J]. 南京邮电大学学报:自然科学版, 2020, 40(5): 112–116. doi: 10.14132/j.cnki.1673-5439.2020.05.012WANG Baoyun and LI Pei. Understanding big data: Irregular structure and graph signal[J]. Journal of Nanjing University of Posts and Telecommunications:Natural Science Edition, 2020, 40(5): 112–116. doi: 10.14132/j.cnki.1673-5439.2020.05.012 [27] CHUNG F R K. Spectral Graph Theory[M]. Providence: American Mathematical Society, 1997: 2–14. [28] 伊鹏飞. 图信号处理技术及OSM数据压缩的研究[D]. [博士论文], 北京理工大学, 2018.YIN Pengfei. The research on graph signal processing and compression of OSM[D]. [Ph. D. dissertation], Beijing Institute of Technology, 2018. [29] 池源. 基于图信号处理的空时信号分布式在线重构算法[D]. [硕士论文], 桂林电子科技大学, 2021.CHI Yuan. Distributed online reconstruction algorithms for spatiotemporal signals based on graph signal processing[D]. [Master dissertation], Guilin University of Electronic Technology, 2021. [30] 杨立山. 图信号采样与重建研究[D]. [博士论文], 北京邮电大学, 2018.YANG Lishan. Research on sampling and reconstruction for graph signals[D]. [Ph. D. dissertation], Beijing University of Posts and Telecommunications, 2018. [31] 汪芬. 图信号高效采样方法研究[D]. [博士论文], 西安电子科技大学, 2021.WANG Fen. Research on efficient subset sampling of graph signals[D]. [Ph. D. dissertation], Xidian University, 2021. [32] O'SHEA K and NASH R. An introduction to convolutional neural networks[J]. arXiv: 1511.08458, 2015. [33] BRONSTEIN M M, BRUNA J, LECUN Y, et al. Geometric deep learning: Going beyond euclidean data[J]. IEEE Signal Processing Magazine, 2017, 34(4): 18–42. doi: 10.1109/msp.2017.2693418 [34] BRUNA J, ZAREMBA W, SZLAM A, et al. Spectral networks and locally connected networks on graphs[C]. The 2nd International Conference on Learning Representations, Banff, Canada, 2013. [35] DEFFERRARD M, BRESSON X, and VANDERGHEYNST P. Convolutional neural networks on graphs with fast localized spectral filtering[C]. The 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, 2016: 3844–3852. [36] KIPF T N and WELLING M. Semi-supervised classification with graph convolutional networks[C]. Proceedings of the 5th International Conference on Learning Representations, Toulon, France, 2017. [37] SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61–80. doi: 10.1109/TNN.2008.2005605 [38] VELIČKOVIĆ P, CUCURULL G, CASANOVA A, et al. Graph attention networks[C]. The 6th International Conference on Learning Representations, Vancouver, Canada, 2018. [39] PAN Shirui, HU Ruiqi, LONG Guodong, et al. Adversarially regularized graph autoencoder for graph embedding[C]. The 27th International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 2018: 2609–2615. [40] KOLDA T G, PINAR A, PLANTENGA T, et al. A scalable generative graph model with community structure[J]. SIAM Journal on Scientific Computing, 2014, 36(5): C424–C452. doi: 10.1137/130914218 [41] LI Yaguang, YU R, SHAHABI C, et al. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting[C]. The 6th International Conference on Learning Representations, Vancouver, Canada, 2018. [42] YAN Shuicheng, XU Dong, ZHANG Benyu, et al. Graph embedding and extensions: A general framework for dimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(1): 40–51. doi: 10.1109/TPAMI.2007.250598 [43] LUNGA D, PRASAD S, CRAWFORD M M, et al. Manifold-learning-based feature extraction for classification of hyperspectral data: A review of advances in manifold learning[J]. IEEE Signal Processing Magazine, 2014, 31(1): 55–66. doi: 10.1109/MSP.2013.2279894 [44] ZOU Jinyi, LI Wei, and DU Qian. Sparse representation-based nearest neighbor classifiers for hyperspectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12): 2418–2422. doi: 10.1109/LGRS.2015.2481181 [45] CHEN Xiaochen, WEI Jia, LI Jinhai, et al. Integrating local and global manifold structures for unsupervised dimensionality reduction[C]. 2014 International Joint Conference on Neural Networks, Beijing, China, 2014: 2837–2843. [46] LUO Huiwu, TANG Yuanyan, LI Chunli, et al. Local and global geometric structure preserving and application to hyperspectral image classification[J]. Mathematical Problems in Engineering, 2015, 2015: 917259. doi: 10.1155/2015/917259 [47] WANG Xiaotao and LIU Fang. Weighted low-rank representation-based dimension reduction for hyperspectral image classification[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(11): 1938–1942. doi: 10.1109/LGRS.2017.2743018 [48] HE Xiaofei and NIYOGI P. Locality preserving projections[C]. The 16th International Conference on Neural Information Processing Systems, Whistler, Canada, 2003: 153–160. [49] QIAN Yuntao and YE Minchao. Hyperspectral imagery restoration using nonlocal spectral-spatial structured sparse representation with noise estimation[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013, 6(2): 499–515. doi: 10.1109/JSTARS.2012.2232904 [50] CHANG Yi, YAN Luxin, ZHAO Xile, et al. Weighted low-rank tensor recovery for hyperspectral image restoration[J]. IEEE Transactions on Cybernetics, 2020, 50(11): 4558–4572. doi: 10.1109/TCYB.2020.2983102 [51] LU Xiaoqiang, WANG Yulong, and YUAN Yuan. Graph-regularized low-rank representation for destriping of hyperspectral images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2013, 51(7): 4009–4018. doi: 10.1109/TGRS.2012.2226730 [52] LIU Na, LI Wei, TAO Ran, et al. Multi-graph-based low-rank tensor approximation for hyperspectral image restoration[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5530314. doi: 10.1109/TGRS.2022.3177719 [53] ZHANG Kai, WANG Min, YANG Shuyuan, et al. Spatial–spectral-graph-regularized low-rank tensor decomposition for multispectral and hyperspectral image fusion[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(4): 1030–1040. doi: 10.1109/JSTARS.2017.2785411 [54] BU Yuanyang, ZHAO Yongqiang, XUE Jize, et al. Hyperspectral and multispectral image fusion via graph Laplacian-guided coupled tensor decomposition[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(1): 648–662. doi: 10.1109/TGRS.2020.2992788 [55] LIU Na, LI We, and TAO Ran. Geometric low-rank tensor approximation for remotely sensed hyperspectral and multispectral imagery fusion[C]. 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, 2022: 2819–2823. [56] VADDI R and MANOHARAN P. Hyperspectral image classification using CNN with spectral and spatial features integration[J]. Infrared Physics & Technology, 2020, 107: 103296. doi: 10.1016/j.infrared.2020.103296 [57] QIN Anyong, SHANG Zhaowei, TIAN Jinyu, et al. Spectral–spatial graph convolutional networks for semisupervised hyperspectral image classification[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(2): 241–245. doi: 10.1109/LGRS.2018.2869563 [58] WAN Sheng, GONG Chen, ZHONG Ping, et al. Multiscale dynamic graph convolutional network for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(5): 3162–3177. doi: 10.1109/TGRS.2019.2949180 [59] HONG Danfeng, GAO Lianru, YAO Jing, et al. Graph convolutional networks for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(7): 5966–5978. doi: 10.1109/TGRS.2020.3015157 [60] WAN Sheng, GONG Chen, ZHONG Ping, et al. Hyperspectral image classification with context-aware dynamic graph convolutional network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(1): 597–612. doi: 10.1109/TGRS.2020.2994205 [61] MOU Lichao, LU Xiaoqiang, LI Xuelong, et al. Nonlocal graph convolutional networks for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(12): 8246–8257. doi: 10.1109/TGRS.2020.2973363 [62] LIU Qichao, XIAO Liang, YANG Jingxiang, et al. CNN-enhanced graph convolutional network with pixel-and superpixel-level feature fusion for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(10): 8657–8671. doi: 10.1109/TGRS.2020.3037361 [63] HE Xin, CHEN Yushi, and GHAMISI P. Dual graph convolutional network for hyperspectral image classification with limited training samples[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5502418. doi: 10.1109/TGRS.2021.3061088 [64] STANKOVIC L, MANDIC D, DAKOVIC M, et al. Graph signal processing--Part I: Graphs, graph spectra, and spectral clustering[J]. arXiv: 1907.03467, 2019. [65] ZENG Hao, LIU Qingjie, ZHANG Mingming, et al. Semi-supervised hyperspectral image classification with graph clustering convolutional networks[J]. arXiv: 2012.10932, 2020. [66] XI Bobo, LI Jiaojiao, LI Yunsong, et al. Semisupervised cross-scale graph prototypical network for hyperspectral image classification[J]. IEEE Transactions on Neural Networks and Learning Systems, To be published. [67] LI Yunsong, XI Bobo, LI Jiaojiao, et al. SGML: A symmetric graph metric learning framework for efficient hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 609–622. doi: 10.1109/JSTARS.2021.3135548 [68] Grupo de Inteligencia Computacional (GIC). Hyperspectral remote sensing scenes[EB/OL]. https://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes, 2021. -
图(7) / 表(1)
计量
- 文章访问数: 1218
- HTML全文浏览量: 569
- PDF下载量: 274
- 被引次数: 0
下载:
