Spatial-spectral collaborative multi-scale vertex component analysis for hyperspectral image endmember extraction

doi:10.11947/j.AGCS.2022.20210718

Abstract

Abstract: Current endmember extraction methods cannot accurately extract the endmembers of complicated ground features, and therefore this paper proposed a spatial-spectral collaborative multi-scale vertex component analysis (VCA) method. Hyperspectral images are firstly jointly clustered and segmented based on multi-feature fusion using spectral features, texture features, and shape features, which makes full use of the spatial heterogeneity information of ground features. Then, multi-scale low-rank matrix decomposition is used to decompose the segmented images and reduce the influence of noise on endmember extraction. Meanwhile, VCA is used to extract endmembers from low-resolution images, coordinate mapping is implemented to search these endmembers of high-resolution images, and vertex component analysis is used to extract endmember from low resolution image. After that, coordinate mapping is used to ferret about the corresponding endmembers in the high-resolution image, and the spectral angle between them is calculated to help accurately decide the pure endmembers. Finally, the proposed method is traversed into all segmented images to obtain the final pure endmembers. The proposed method is verified experimentally by using simulated and real GF-5 hyperspectral data. Experimental results show that the CVCA method can extract high-precision pure endmembers and has high calculation efficiency.

Key words: endmember, hyperspectral, spatial-spectral collaboration, vertex component analysis

CLC Number:

P227

SUN Weiwei, CHANG Minghui, MENG Xiangchao, YANG Gang, REN Kai. Spatial-spectral collaborative multi-scale vertex component analysis for hyperspectral image endmember extraction[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(4): 587-598.

References

[1] 孙伟伟,杨刚,陈超,等. 中国地球观测遥感卫星发展现状及文献分析[J]. 遥感学报,2020,24(5): 479-510. SUN Weiwei,YANG Gang,CHEN Chao,et al. Development status and literature analysis of China's earth observation and remote sensing satellites [J]. National Remote Sensing Bulletin,2020,24 (5): 479-510.
[2] 孟祥超,孙伟伟. 基于多分辨率分析的GF-5和GF-1遥感影像空—谱融合[J]. 遥感学报,2020, 24(4): 379-387. MENG Xiangchao,SUN Weiwei. Spatial spectral fusion of GF-5 and GF-1 remote sensing images based on multi-resolution analysis [J]. National Remote Sensing Bulletin,2020,24 (4): 379-387.
[3] SUN Weiwei,MA Jun,YANG Gang. A Poisson nonnegative matrix factorization method with parameter subspace clustering constraint for endmember extraction in hyperspectral imagery[J]. ISPRS Journal of Photogrammetry Remote Sensing,2017(128): 27-39.
[4] 韩竹,高连如,张兵,等. 高分五号高光谱图像自编码网络非线性解混[J]. 遥感学报,2019,24(4): 388-400. HAN Zhu,GAO Lianru,ZHANG Bing,et al.Nonlinear unmixing of GF-5 hyperspectral image self coding network [J].National Remote Sensing Bulletin,2019,24 (4): 388-400.
[5] XU Xiang,LI Jun,WU Changshan,et al. Regional clustering -based spatial preprocessing for hyperspectral unmixing[J]. Remote Sensing of Environment,2018(204): 333-346.
[6] 丁海勇,史文中. 利用卡方分布改进N-FINDR端元提取算法[J]. 遥感学报,2012,17(1): 122-137. DING Haiyong,SHI Wenzhong.Improved N-FINDR endmember extraction algorithm using chi square distribution [J]. National Remote Sensing Bulletin,2012,17 (1): 122-137.
[7] 田玉刚,杨贵. 端元快速提取的光谱梯度特征搜索法[J]. 测绘学报,2015, 44(2): 214-219,227. DOI: 10.11947/j.AGCS.2015.20130392. TIAN Yugang,YANG Gui.Spectral gradient feature search method for rapid extraction of endmembers [J]. Acta Geodaetica et Cartographica Sinica,2015,44 (2): 214-219, 227. DOI: 10.11947/j.AGCS.2015.20130392.
[8] 王丽,王威,刘勃妮.顶点成分分析法应用于高光谱图像端元提取[J].信息技术,2022(1):37-43.DOI:10.13274/j.cnki.hdzj.2022.01.008. WANG Li,WANG Wei,LIU Boni. Application of vertex component analysis to endmember extraction of hyperspectral images [J]. Information Technology,2022(1): 37-43. DOI:10.13274/ j.cnki. hdzj. 2022.01.008.
[9] 袁博. 基于混合像元空间与谱间相关性模型的NMF线性盲解混[J]. 测绘学报,2019,48(9): 1151-1160.DOI: 10.11947/j.AGCS.2019.20180054. YUAN Bo. NMF linear blind unmixing based on the spatial and spectral correlation model of mixed pixels [J]. Acta Geodaetica et Cartographica Sinica,2019,48 (9): 1151-1160. DOI: 10.11947/j.AGCS.2019.20180054.
[10] 徐君,王彩玲,王丽. 采用PPI算法改进的一种数学形态学端元提取方法 [J]. 测绘学报,2019, 48(8): 996-1003.DOI:10.11947/j.AGCS.2019.20180475. XU Jun,WANG Cailing, WANG Li. A mathematical morphology endmember extraction method improved by PPI algorithm [J]. Acta Geodaetica et Cartographica Sinica,2019,48 (8): 996-1003. DOI:10.11947/j.AGCS.2019.20180475.
[11] 李二森,朱书龙,周晓明,等. 高光谱图像端元提取算法研究进展与比较 [J]. 遥感学报,2011,15(4): 659-679. LI Ersen,ZHU Shulong,ZHOU Xiaoming,et al. Research progress and comparison of endmember extraction algorithms for hyperspectral images [J]. National Remote Sensing Bulletin,2011,15 (4): 659-679.
[12] 程熙,沈占锋,骆剑承,等. 利用混合光谱分解与SVM估算不透水面覆盖率[J]. 遥感学报, 2011,15(6): 1235-1247. CHENG Xi,SHEN Zhanfeng,LUO Jiancheng,et al. Estimation of impervious surface coverage using mixed spectral decomposition and SVM [J]. National Remote Sensing Bulletin, 2011,15 (6): 1235-1247.
[13] 方俊龙,郭宝峰,沈宏海,等. 一种改进的基于自动形态学的端元提取算法[J]. 激光技术,2017,41(1): 106-112. FANG Junlong,GUO Baofeng,SHEN Honghai,et al. An improved endmember extraction algorithm based on automatic morphology [J]. Laser Technology,2017,41 (1): 106-112.
[14] 郭诗韵. 一种基于空谱协同的高光谱端元提取方法研究[J]. 地理空间信息,2020,18(2): 34-37. GUO Shiyun. Research on a hyperspectral endmember extraction method based on spatial spectrum cooperation [J]. Geospatial Information, 2020,18 (2): 34-37.
[15] 王忠美,杨晓梅,顾行发. 张量组稀疏表示的高光谱图像去噪算法[J]. 测绘学报, 2017,46(5): 614-622.DOI: 10.11947/j.AGCS.2017.20150403. WANG Zhongmei,YANG Xiaomei, GU Xingfa. Hyperspectral Image denoising based on tensor group sparse representation[J]. Acta Geodaetica et Cartographica Sinica,2017,46(5): 614-622. DOI: 10.11947/j.AGCS.2017.20150403.
[16] EKLUNDH L,SINGH A. A comparative analysis of standardised and unstandardised principal components analysis in remote sensing[J]. International Journal of Remote Sensing,1993,14(7): 1359-1370.
[17] XU Qian,VARADARAJAN S,CHAKRABARTI C,et al. A distributed canny edge detector: algorithm and FPGA implementation[J]. IEEE Transactions on Image Processing,2014,23(7): 2944-2960.
[18] SATPATHY A,XU Dongjiang,ENG H-L. LBP-based edge-texture features for object recognition[J]. IEEE Transactions on Image Processing,2014,23(5): 1953-1964.
[19] SHAH D,ZAVERI T. Hyperspectral endmember extraction using band quality[C]//Proceedings of 2019 IEEE India Council International Conference. Rajkot, India: IEEE, 2019: 1-4.
[20] SHAH D,ZAVERI T,TRIVEDI Y N,et al. Entropy-based convex set optimization for spatial-spectral endmember extraction from hyperspectral images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2020,13(8):4200-4213.
[21] KUMAR U,CRISTINA M,NEMANI R,et al. Sparse unmixing via variable splitting and augmented Lagrangian for vegetation and urban area classification using Landsat data[J].ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2015, 40(7):59-65.