Unsupervised band selection for hyperspectral image classification using the Wasserstein metric-based configuration entropy

doi:10.11947/j.AGCS.2021.20200006

Abstract

Abstract: Band selection relies on the quantification of band information. Conventional measurements such as Shannon entropy only consider the composition information (e.g., types and ratios of pixels) but ignore the configuration information (e.g., the spatial distribution of pixels). The latter could be quantified by Boltzmann entropy. Among all the metrics of Boltzmann entropy, the Wasserstein metric-based configuration entropy (Wasserstein entropy for short) removes the redundant information of the continuous pixels. However, it is limited to 4-neighborhood. This article improves it to 8-neighborhood. Taking the hyperspectral images of Indian Pines and Italian Pavia University as examples, we used the difference of Wasserstein entropy to measure band correlation and then employed the unsupervised sub-optimal searching algorithm to determine the optimal band combination. We used the support vector machine classifier for image classification. Finally, we compared the accuracy of image classification based on the difference of Wasserstein entropy, mutual information, four types of normalized mutual information, and two variants of relative entropy. Results show that both the 4-neighborhood and 8-neighborhood Wasserstein entropy can be used for band selection of hyperspectral images, especially when few bands are considered. The 8-neighborhood Wasserstein entropy works better than 4-neighborhood.

Key words: hyperspectral image, image classification, Shannon entropy, Wasserstein configuration entropy, band selection

CLC Number:

P237

ZHANG Hong, WU Zhiwei, WANG Jicheng, GAO Peichao. Unsupervised band selection for hyperspectral image classification using the Wasserstein metric-based configuration entropy[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(3): 405-415.

References

[1] 黄鸿, 郑新磊. 高光谱影像空-谱协同嵌入的地物分类算法[J]. 测绘学报, 2016, 45(8):964-972. DOI:10.11947/j.AGCS.2016.20150654. HUANG Hong, ZHENG Xinlei. Hyperspectral image land cover classification algorithm based on spatial-spectral coordination embedding[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(8):964-972. DOI:10.11947/j.AGCS.2016.20150654.
[2] CASA R, UPRETI D, PALOMBO A, et al. Evaluation and exploitation of retrieval algorithms for estimating biophysical crop variables using Sentinel-2, Venus, and PRISMA satellite data[J]. Journal of Geodesy and Geoinformation Science, 2020, 3(4):79-88. DOI:10.11947/j.JGGS.2020.0408.
[3] 杨钊霞, 邹峥嵘, 陶超, 等. 空-谱信息与稀疏表示相结合的高光谱遥感影像分类[J]. 测绘学报, 2015, 44(7):775-781. DOI:10.11947/j.AGCS.2015.20140207. YANG Zhaoxia, ZOU Zhengrong, TAO Chao, et al. Hyperspectral image classification based on the combination of spatial-spectral feature and sparse representation[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(7):775-781. DOI:10.11947/j.AGCS.2015.20140207.
[4] SUN Weiwei, DU Qian. Graph-regularized fast and robust principal component analysis for hyperspectral band selection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(6):3185-3195.
[5] GAO Peichao, WANG Jicheng, ZHANG Hong, et al. Boltzmann entropy-based unsupervised band selection for hyperspectral image classification[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(3):462-466.
[6] SAWANT S S, MANOHARAN P. Unsupervised band selection based on weighted information entropy and 3D discrete cosine transform for hyperspectral image classification[J]. International Journal of Remote Sensing, 2020, 41(10):3948-3969.
[7] CHANG C I, DU Qian, SUN T L, et al. A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(6):2631-2641.
[8] CHANG C I, WANG Su. Constrained band selection for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(6):1575-1585.
[9] MARTÍNEZ-USÓMARTINEZ-USO A, PLA F, SOTOCA J M, et al. Clustering-based hyperspectral band selection using information measures[J]. IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(12):4158-4171.
[10] GONG Jianya, JI Shunping. Photogrammetry and deep learning[J]. Journal of Geodesy and Geoinformation Science, 2018, 1(1):1-15. DOI:10.11947/j.JGGS.2018.0101.
[11] ZENG Meng, CAI Yaoming, CAI Zhihua, et al. Unsupervised hyperspectral image band selection based on deep subspace clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(12):1889-1893.
[12] 曾梦, 宁彬, 蔡之华, 等. 使用深度对抗子空间聚类实现高光谱波段选择[J]. 计算机应用, 2020, 40(2):381-385. ZENG Meng, NING Bin, CAI Zhihua, et al. Hyperspectral band selection based on deep adversarial subspace clustering[J]. Journal of Computer Applications, 2020, 40(2):381-385.
[13] CAI Yaoming, LIU Xiaobo, CAI Zhihua. BS-Nets:an end-to-end framework for band selection of hyperspectral image[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(3):1969-1984.
[14] 严阳, 华文深, 刘恂, 等. 基于高光谱基本准则的波段选择方法[J]. 光学技术, 2018, 44(5):634-640. YAN Yang, HUA Wenshen, LIU Xun, et al. Band selection method based on hyperspectral fundamental criterion[J]. Optical Technique, 2018, 44(5):634-640.
[15] SHANNON C E. A mathematical theory of communication[J]. The Bell System Technical Journal, 1948, 27(3):379-423.
[16] 李志林, 刘启亮, 高培超. 地图信息论:从狭义到广义的发展回顾[J]. 测绘学报, 2016, 45(7):757-767. DOI:10.11947/j.AGCS.2016.20160235. LI Zhilin, LIU Qiliang, GAO Peichao. Entropy-based cartographic communication models:evolution from special to general cartographic information theory[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(7):757-767. DOI:10.11947/j.AGCS.2016.20160235.
[17] LI Zhilin, HUANG Peizhi. Quantitative measures for spatial information of maps[J]. International Journal of Geographical Information Science, 2002, 16(7):699-709.
[18] CAO Xianghai, HAN Jungong, YANG Shuyuan, et al. Band selection and evaluation with spatial information[J]. International Journal of Remote Sensing, 2016, 37(19):4501-4520.
[19] CUSHMAN S A. Calculating the configurational entropy of a landscape mosaic[J]. Landscape Ecology, 2016, 31(3):481-489.
[20] GAO Peichao, ZHANG Hong, LI Zhilin. A hierarchy-based solution to calculate the configurational entropy of landscape gradients[J]. Landscape Ecology, 2017, 32(6):1133-1146.
[21] ZHAO Yuan, ZHANG Xinchang. Calculating spatial configurational entropy of a landscape mosaic based on the Wasserstein metric[J]. Landscape Ecology, 2019, 34(8):1849-1858.
[22] SUKHOV V I. Information capacity of a map entropy[J]. Geodesy and Aerophotography, 1967, 10(4):212-215.
[23] HARALICK R M, SHANMUGAM K, DINSTEIN I H. Textural features for image classification[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973(6):610-621.
[24] BOLTZMANN L. Weitere studien über das wärmegleichge-wicht unter gasmolekülen[M]. Sitzungsberichte Akademie der Wissenschaften, 1872, 66:275-370.
[25] ZHANG Mingyang, GONG Maoguo, MAO Yishun, et al. Unsupervised feature extraction in hyperspectral images based on Wasserstein generative adversarial network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(5):2669-2688.
[26] ESTEVEZ P A, TESMER M, PEREZ C A, et al. Normalized mutual information feature selection[J]. IEEE Transactions on Neural Networks, 2009, 20(2):189-201.
[27] VINH N X, EPPS J, BAILEY J. Information theoretic measures for clusterings comparison:Variants, properties, normalization and correction for chance[J]. Journal of Machine Learning Research, 2010, 11(1):2837-2854.
[28] KULLBACK S, LEIBLER R A. On information and sufficiency[J]. Annals of Mathematical Statistics, 1951, 22(1):79-86.
[29] 邬建国. 景观生态学——格局、过程、尺度与等级[M]. 2版. 北京:高等教育出版社, 2007:125-147. WU Jianguo. Landscape ecology:pattern, process, scale and hierarchy[M]. 2nd ed. Beijing:Higher Education Press, 2007:125-147.
[30] HENEBRY G M. Spatial model error analysis using autocorrelation indices[J]. Ecological Modelling, 1995, 82(1):75-91.
[31] BAUMGARDNER M F, BIEHL L L, LANDGREBE D A. 220 band aviris hyperspectral image data set:June 12, 1992 Indian pine test site 3[J]. Purdue University Research Repository, 2015.
[32] GEGE P, BERAN D, MOOSHUBER W, et al. System analysis and performance of the new version of the imaging spectrometer ROSIS[C]//Proceedings of the 1st EARSel Workshop on Imaging Spectroscopy Remote Sensing Laboratories. 1998:29-35.
[33] 魏立飞, 余铭, 钟燕飞, 等. 空-谱融合的条件随机场高光谱影像分类方法[J]. 测绘学报, 2020, 49(3):343-354. DOI:10.11947/j.AGCS.2020.20190042. WEI Lifei, YU Ming, ZHONG Yanfei, et al. Hyperspectral image classification method based on space-spectral fusion conditional random field[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(3):343-354. DOI:10.11947/j.AGCS.2020.20190042.