结合空谱结构与改进局部密度的高光谱图像波段选择方法

翁谦; 安远; 陈光剑; 吴瑞姣; 林嘉雯

doi:10.11834/jrs.20243227

浏览量 : 0 下载量: 0 CSCD: 0 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

结合空谱结构与改进局部密度的高光谱图像波段选择方法
Integration of Spatial Spectral Structure with Improved Local Density for Hyperspectral Image Band Selection Method
2024年页码：1-20
网络出版日期： 2024-04-01 ，
DOI： 10.11834/jrs.20243227

扫描看全文

翁谦，安远，陈光剑，吴瑞姣，林嘉雯.XXXX.结合空谱结构与改进局部密度的高光谱图像波段选择方法.遥感学报，XX（XX）： 1-20

WEN Qian，AN Yuan，CHEN Guangjian，WU Ruijiao，LIN Jiawen. XXXX. Integration of Spatial Spectral Structure with Improved Local Density for Hyperspectral Image Band Selection Method. National Remote Sensing Bulletin， XX（XX）：1-20
翁谦，安远，陈光剑，吴瑞姣，林嘉雯.XXXX.结合空谱结构与改进局部密度的高光谱图像波段选择方法.遥感学报，XX（XX）： 1-20 DOI： 10.11834/jrs.20243227.

WEN Qian，AN Yuan，CHEN Guangjian，WU Ruijiao，LIN Jiawen. XXXX. Integration of Spatial Spectral Structure with Improved Local Density for Hyperspectral Image Band Selection Method. National Remote Sensing Bulletin， XX（XX）：1-20 DOI： 10.11834/jrs.20243227.

摘要

波段选择是高光谱遥感图像降维的一项重要任务，其目标是选择包含较少冗余信息、较大信息量和具有类别可分性的波段子集。为解决基于近邻子空间划分的波段选择方法没有考虑地物空间分布和计算聚类中心时忽略噪声波段影响的问题，本文提出了一种结合空谱结构与改进局部密度的高光谱图像波段选择方法。该方法首先对高光谱图像进行基于熵率的图像分割获得高光谱图像同质区域，综合同质区域相关系数矩阵获得图像区域级近邻波段相关系数向量；其次，用高斯核平滑全局近邻波段相关系数向量以降低噪声波段的影响，并根据极值点进行波段分组；然后，最大化改进局部密度和波段信息熵的乘积作为选取代表性波段的标准；最后，在Indian Pines、Botswana 和Salinas高光谱图像数据集上，通过SVM、KNN和LDA分类器上进行分类实验。结果表明：①对比像素级相关系数划分方法，利用区域级相关系数使得近邻波段分组更为合理，降低波段冗余性，同时还保留了部分潜在特征波段，在3个数据集上的分类性能分别提高了2.63%，0.68%，0.16%；②对比仅使用信息熵的波段衡量方法，本文提出的最大化改进的局部密度和信息熵乘积的方法是有效的，在3个数据集上OA分别提高了4.13%、0.5%和0.21%；③对比其他6种先进波段选择方法，本文方法在3个数据集上的OA分别从62.34%提高到75.03%、86.74%提高到88.28%和86.04%提高到92.36%。此外，所选择的波段子集在分布上较为分散，主要集中在信息熵较高的区域，同时避免了选择噪声波段。综上，本文提出的波段选择方法所得波段子集具有较低的冗余性、丰富的信息量、强的类别可分性，并且对噪声具有较强的鲁棒性，能够有效地解决高光谱图像波段选择的问题。

Abstract

Band selection is a crucial task in the dimensionality reduction of hyperspectral remote sensing imagery. Its objective is to choose a subset of bands that contain less redundant information

higher information content

and exhibit class discriminability. To address the issues of existing band selection methods based on nearest neighbor subspace partitioning

which do not consider the spatial distribution of objects and neglect the impact of noisy bands when computing cluster centers

this paper proposes a hyperspectral image band selection method that integrates spatial spectral structure with improved local density

referred to as ISSS-ELD.This method first performs image segmentation on the hyperspectral image using entropy-based approach to acquire homogeneous regions. The composite region-level neighboring band correlation coefficient vector is obtained by integrating the correlation coefficient matrix of these homogeneous regions. Subsequently

a Gaussian kernel is applied to globally smooth the neighboring band correlation coefficient vector

reducing the influence of noisy bands. Bands are grouped based on extremum points in the smoothed vector. Finally

the product of the maximized improved local density and band information entropy serves as the criterion for selecting representative bands.This paper conducted experiments on hyperspectral image datasets including Indian Pines

Botswana

and Salinas. Different band selection methods were evaluated by calculating metrics such as classification accuracy

average correlation coefficient

and noise robustness of the selected bands. The results indicate that: (1) Compared to pixel-level correlation-based partitioning methods

the utilization of region-level correlation coefficients results in more reasonable grouping of neighboring bands

reducing band redundancy while retaining some potential characteristic bands. The classification performance on the three datasets improved by 2.63%

0.68%

and 0.16% respectively. (2) In contrast to methods solely using information entropy for band assessment

the proposed approach of maximizing the product of improved local density and information entropy proves effective. On the three datasets

Overall Accuracy (OA) increased by 4.13%

0.5%

and 0.21% respectively. (3) Compared to six other advanced band selection methods

the proposed method achieved significant performance improvements: OA increased from 62.34% to 75.03%

from 86.74% to 88.28%

and from 86.04% to 92.36% on the three datasets. Furthermore

the selected subset of bands by our method is dispersed

concentrating in regions with higher information entropy

and effectively avoiding the inclusion of noisy bands.In summary

the band subset selected by the proposed band selection method exhibits low redundancy

high information content

strong class separability

and robustness against noise

effectively addressing the challenges in hyperspectral image band selection.

关键词

遥感高光谱图像波段选择子空间划分峰值密度局部密度信息熵分类

Keywords

remote sensinghyperspectral imagesband selectionsubspace divisionpeak densitylocal densityinformation entropyclassification

references

Cai Y, Zhang Z, Liu X and Cai Z. 2020. Efficient Graph Convolutional Self-Representation for Band Selection of Hyperspectral Image, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13(99): 1-1 [DOI: 10.1109/JSTARS.2020.3018229http://dx.doi.org/10.1109/JSTARS.2020.3018229]

Chang C I, Du Q, Sun T L and Althouse M. 1999. A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 37(6): 2631-2641 [DOI: 10.1109/36.803411http://dx.doi.org/10.1109/36.803411]

Chang C I, Wu C C, Liu K H, Chen H M, Chen C and Wen C H. 2015. Progressive band processing of linear spectral unmixing for hyperspectral imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(6): 2583-2597 [DOI: 10.1109/JSTARS.2014.2371438http://dx.doi.org/10.1109/JSTARS.2014.2371438]

Du Q and Yang H. 2008. Similarity-based unsupervised band selection for hyperspectral image analysis. IEEE Geoscience and Remote Sensing Letters, 5(4): 564-568 [DOI: 10.1109/LGRS.2008.2000619http://dx.doi.org/10.1109/LGRS.2008.2000619]

Gong M, Zhang M and Yuan Y. 2016. Unsupervised band selection based on evolutionary multiobjective optimization for hyperspectral images. IEEE Transactions on Geoscience and Remote Sensing, 54(1): 544-557 [DOI: 10.1109/TGRS.2015.2461653http://dx.doi.org/10.1109/TGRS.2015.2461653]

Govender M, Chetty K and Bulcock H. 2007. A review of hyperspectral remote sensing and its application in vegetation and water resource studies. Water S. A., 33(2): 145-151 [DOI: 10.4314/wsa.v33i2.49049http://dx.doi.org/10.4314/wsa.v33i2.49049]

Groves P and Bajcsy P. 2003. Methodology for hyperspectral band and classification model selection. MD, USA: IEEE 120-128 [DOI: 10.1109/WARSD.2003.1295183http://dx.doi.org/10.1109/WARSD.2003.1295183]

Huang R and Li X H. 2008. Band selection based on evolution algorithm and sequential search for hyperspectral classification. Shanghai, China: IEEE: 1270-1273 [DOI: 10.1109/ICALIP.2008.4590145http://dx.doi.org/10.1109/ICALIP.2008.4590145]

Jia S, Tang G, Zhu J and Li Q. 2016. A novel ranking-based clustering approach for hyperspectral band selection. IEEE transactions on geoscience and remote sensing, 54(1): 88-102 [DOI: 10.1109/TGRS.2015.2450759http://dx.doi.org/10.1109/TGRS.2015.2450759]

Jia S, Yuan Y, Li N, Liao J, Huang Q, Jia X and Xu M. 2022. A multiscale superpixel-level group clustering framework for hyperspectral band selection. IEEE Transactions on Geoscience and Remote Sensing, 60(1): 1-18 [DOI: 10.1109/TGRS.2022.3150361http://dx.doi.org/10.1109/TGRS.2022.3150361]

Kumar B, Dikshit O, Gupta A and Singh M K. 2020. Feature extraction for hyperspectral image classification: a review. International Journal of Remote Sensing, 41(16): 6248-6287 [DOI: 10.1080/01431161.2020.1736732http://dx.doi.org/10.1080/01431161.2020.1736732]

Li F and Lu H C. 2016. Hyperspectral images band selection using multi-dictionary based sparse representation. IEEE International Geoscience and Remote Sensing Symposium. Beijing, China: IEEE: 2769-2772 [DOI: 10.1109/IGARSS.2016.7729715http://dx.doi.org/10.1109/IGARSS.2016.7729715]

Li J M and Qian Y T. 2011. Clustering-based hyperspectral band selection using sparse nonnegative matrix factorization. Journal of Zhejiang University-Science C-Computers & Electronics, 12(7): 542-549 [DOI: 10.1631/jzus.C1000304http://dx.doi.org/10.1631/jzus.C1000304]

Liu M Y, Tuzel O, Ramalingam S and Chellappa R. 2011. Entropy rate superpixel segmentation. CVPR 2011. CO, USA: IEEE: 2097-2104 [DOI: 10.1109/CVPR.2011.5995323http://dx.doi.org/10.1109/CVPR.2011.5995323]

Liu X S, Ge L, Wang B and Zhang L M. 2012. An unsupervised band selection algorithm for hyperspectral imagery based on maximal information. Journal of Infrared and Millimeter Waves, 31(2): 166-171

刘雪松, 葛亮, 王斌, 张立明. 2012. 基于最大信息量的高光谱遥感图像无监督波段选择方法. 红外与毫米波学报, 31(2): 166-171 [DOI: 10.3724/SP.J.1010.2012.00166http://dx.doi.org/10.3724/SP.J.1010.2012.00166]

Lu G and Fei B. 2014. Medical hyperspectral imaging: a review. Journal of Biomedical Optics, 19(1): 10901 [DOI: 10.1117/1.JBO.19.1.010901http://dx.doi.org/10.1117/1.JBO.19.1.010901]

Lu Y，Ren Y and Cui B G. 2022. Noise robust band selection method for hyperspectral images. National Remote Sensing Bulletin, 26（11）：2382-2398

路燕, 任月 and 崔宾阁. 2022. 噪声鲁棒的高光谱图像波段选择方法. 遥感学报, 26(11): 2382-2398 [DOI: 10.11834/jrs.20211128http://dx.doi.org/10.11834/jrs.20211128]

Martinez-Uso A, Pla F, Sotoca J M and Garcia-Sevilla P. 2007. Clustering-based hyperspectral band selection using in-formation measures. IEEE transactions on geoscience and remote sensing, 45(12): 4158-4171 [DOI: 10.1109/TGRS.2007.904951http://dx.doi.org/10.1109/TGRS.2007.904951]

Mehmood R, Dawood H, Bie R and Ahmad H. 2015. Fuzzy clustering by fast search and find of density peaks. 2015 International Conference on Identification, Information, and Knowledge in the Internet of Things (IIKI). Beijing, China: IEEE: 258-261 [DOI: 10.1109/IIKI.2015.62http://dx.doi.org/10.1109/IIKI.2015.62]

Morales G, Sheppard J W, Logan R D and Shaw J A. 2021. Hyperspectral dimensionality reduction based on inter-band redundancy analysis and greedy spectral selection. Remote Sensing, 13(18): 3649-3652 [DOI: 10.3390/rs13183649http://dx.doi.org/10.3390/rs13183649]

Patro R N, Subudhi S, Biswal P K and Dell A F. 2021. A review on unsupervised band selection techniques: land cover classification for hyperspectral earth observation data. IEEE Geoscience and Remote Sensing Magazine, 9(3) 72-111 [DOI: 10.1109/MGRS.2021.3051979http://dx.doi.org/10.1109/MGRS.2021.3051979]

Qian Y, Yao F and Jia S. 2008. Band selection for hyperspectral imagery using affinity propagation. Computer Vision Iet, 3(4): 213-222 [DOI: 10.1049/iet-cvi.2009.0034http://dx.doi.org/10.1049/iet-cvi.2009.0034]

Shahshahani B M and Landgrebe D A. 1994. The effect of unlabeled samples in reducing the small sample size problem and mitigating the hughes phenomenon. IEEE Transactions on Geoscience and Remote Sensing, 32(5): 1087-1095 [DOI: 10.1109/36.312897http://dx.doi.org/10.1109/36.312897]

Shannon C E. 1948. A mathematical theory of communication, 1948. Bell System Technical Journal, 27(4): 623-656 [DOI: 10.1002/j.1538-7305.1948.tb01338.xhttp://dx.doi.org/10.1002/j.1538-7305.1948.tb01338.x]

Shimoni M, Haelterman R and Perneel C. 2019. Hypersectral imaging for military and security applications: combining myriad processing and sensing techniques. IEEE Geoscience and Remote Sensing Magazine, 7(2): 101-117 [DOI: 10.1109/MGRS.2019.2902525http://dx.doi.org/10.1109/MGRS.2019.2902525]

Sun W and Du Q. 2019. Hyperspectral band selection: a review. IEEE Geoscience and Remote Sensing Magazine, 7(2): 118-139 [DOI: 10.1109/MGRS.2019.2911100http://dx.doi.org/10.1109/MGRS.2019.2911100]

Sun W, Yang G, Peng J, Meng X and Du Q. 2021. A multiscale spectral features graph fusion method for hyperspectral band selection. IEEE Transactions on Geoscience and Remote Sensing, 60(8): 1-12 [DOI: 10.1109/TGRS.2021.3102246http://dx.doi.org/10.1109/TGRS.2021.3102246]

Sun W W，Yang G，Peng J T and Meng X C. 2022. Robust multi-feature spectral clustering for hyperspectral band selection. National Remote Sensing Bulletin， 26（2）：397-405

孙伟伟, 杨刚, 彭江涛 and 孟祥超. 2022. 鲁棒多特征谱聚类的高光谱图像波段选择. 遥感学报, 26(2): 397-405 [DOI: 10.11834/jrs.20209165http://dx.doi.org/10.11834/jrs.20209165]

Tang Chang,Wang Jun.A Hyperspectral Band Selection Method via Adjacent Subspace Partition[J].Journal of Tianjin University(Science and Technology), 2022(3):0255-0262

唐厂, 王俊. 2022. 基于近邻子空间划分的高光谱图像波段选择方法. 天津大学学报, 055(003): 255-262 [DOI: 10.11784/tdxbz202012004http://dx.doi.org/10.11784/tdxbz202012004]

Wang Q, Li Q and Li X. 2019. Hyperspectral band selection via adaptive subspace partition strategy. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12(12): 4940-4950 [DOI: 10.1109/JSTARS.2019.2941454http://dx.doi.org/10.1109/JSTARS.2019.2941454]

Wang Q, Li Q and Li X. 2020a. A fast neighborhood grouping method for hyperspectral band selection. IEEE Transac-tions on Geoscience and Remote Sensing, 59(06): 5028-5039 [DOI: 10.1109/TGRS.2020.3011002http://dx.doi.org/10.1109/TGRS.2020.3011002]

Wang X, Wang G and Michel C. 2021. Hyperspectral band selection based on adaptive neighborhood grouping and local structure correlation. Journal of Sensors, 2021(1): 21-21 [DOI: 10.1155/2021/5530385http://dx.doi.org/10.1155/2021/5530385]

Yang R, Su L, Zhao X, Wan H and Sun J. 2017. Representative band selection for hyperspectral image classification. Journal of Visual Communication and Image Representation, 48(oct.): 396-403 [DOI: 10.1016/j.jvcir.2017.02.002http://dx.doi.org/10.1016/j.jvcir.2017.02.002]

Yang H Y, and Du J M. 2022. Research Progress Review of Hyperspectral Remote Sensing Image Band Selection. Computer Engineering and Application, 58(10): 1-12

杨红艳, 杜健民. 2022. 高光谱遥感图像波段选择研究进展综述. 计算机工程与应用, 58(10): 1-12 [DOI: 10.3778/j.issn.1002-8331.2111-0403http://dx.doi.org/10.3778/j.issn.1002-8331.2111-0403]

Yang H, Du Q and Chen G S. 2012. Particle swarm optimization-based hyperspectral dimensionality reduction for urban land cover classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 5(2): 544-554 [DOI: 10.1109/JSTARS.2012.2185822http://dx.doi.org/10.1109/JSTARS.2012.2185822]

Yu D, Xing Q, Ping S and Yu D. 2013. Estimating Secchi depth by simulated HSI of HJ-1A from in situ hyperspectral data: A case study in Sishili Bay. International Conference on Computer Vision in Remote Sensing. Xiamen, China: IEEE: 291-295 [DOI: 10.1109/CVRS.2012.6421277http://dx.doi.org/10.1109/CVRS.2012.6421277]

Zhang W, Li X, Dou Y and Zhao L. 2018. A geometry-based band selection approach for hyperspectral image analysis. IEEE Transactions on Geoscience and Remote Sensing, 56(8): 4318-4333 [DOI: 10.1109/TGRS.2018.2811046http://dx.doi.org/10.1109/TGRS.2018.2811046]

Zhang Y S, Wang X X, Jiang X W and Zhou Y C. 2022a. Marginalized graph self-representation for unsupervised hyperspectral band selection. IEEE Transactions on Geoscience and Remote Sensing, 60(10): 1-12 [DOI: 10.1109/TGRS.2021.3121671http://dx.doi.org/10.1109/TGRS.2021.3121671]

Zhang Y S, Wang X X, Jiang X W and Zhou Y C. 2022b. Robust dual graph self-representation for unsupervised hyperspectral band selection. IEEE Transactions on Geoscience and Remote Sensing, 60(8): 1-13 [DOI: 10.1109/TGRS.2022.3203207http://dx.doi.org/10.1109/TGRS.2022.3203207]

Zhao L, Wang L G and Liu D F. 2019. A subspace band selection method for hyperspectral imagery. Journal of Remote Sensing, 23(5): 904–910

赵亮, 王立国 and 刘丹凤. 2019. 高光谱图像子空间的波段选择. 遥感学报, 23(5): 904-910 [DOI: 10.11834/jrs.20197508http://dx.doi.org/10.11834/jrs.20197508]

文章被引用时，请邮件提醒。

提交

高光谱图像类别独立的域适应分类

洞庭湖湿地净初级生产力估算研究

玉米大豆生长中后期遥感辨识的指示性特征研究

鲁棒多特征谱聚类的高光谱影像波段选择

面向对象与卷积神经网络模型的GF-6 WFV影像作物分类