Band selection algorithm for reverse nearest neighbor density peak clustering of hyperspectral images

doi:10.11947/j.AGCS.2024.20220350

Abstract

Abstract: The density peak clustering band selection algorithm uses the local density to describe the density information of the band. However, the existing local density is easy to ignore the global information of the band distribution and can't effectively describe the distribution characteristics of the band, resulting in the limited classification accuracy of the band subset. In order to solve the above problems, this paper proposes a density peak clustering band selection algorithm based on reverse nearest neighbor. Firstly, the K-nearest neighbor directed graph is constructed by using the band and its K-nearest neighbor to obtain the reverse nearest neighbor of the band, as well as the shared nearest neighbor and shared reverse nearest neighbor between bands. Then, the union number of shared nearest neighbors and shared reverse nearest neighbors is used as the similarity between bands, and the enhanced local density is constructed by using the average Euclidean distance and similarity between bands and their reverse nearest neighbors. Finally, the product of enhanced local density, distance factor and information entropy is taken as the weight value, and the segment subset is selected according to the weight value. In order to improve the efficiency and practicability of the experiment, an adaptive K value method is also proposed in this paper. The experimental results on three hyperspectral standard data sets show that the band subset obtained by this algorithm has better classification performance than the band selected by other advanced algorithms, especially when the number of bands is small, and the calculation efficiency is high.

Key words: hyperspectral image, band selection, density peak clustering, reverse nearest neighbor, local density, adaptive K value

CLC Number:

P237

SUN Genyun, LI Renren, ZHANG Aizhu, AN Na, FU Hang, PAN Zhaojie. Band selection algorithm for reverse nearest neighbor density peak clustering of hyperspectral images[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(1): 8-19.

References

[1] 童庆禧, 张兵, 张立福. 中国高光谱遥感的前沿进展[J]. 遥感学报, 2016, 20(5): 689-707. TONG Qingxi, ZHANG Bing, ZHANG Lifu. Current progress of hyperspectral remote sensing in China[J]. Journal of Remote Sen-sing, 2016, 20(5): 689-707.
[2] WANG Kexian, ZHENG Shunyi, LI Rui, et al.A deep double-channel dense network for hyperspectral image classifica-tion[J]. Journal of Geodesy and Geoinformation Science, 2021, 4(4): 46-62.
[3] HUGHES G. On the mean accuracy of statistical pattern recognizers[J]. IEEE Transactions on Information Theory, 1968, 14(1): 55-63.
[4] 谷雨, 徐英, 郭宝峰. 融合空谱特征和集成超限学习机的高光谱图像分类[J]. 测绘学报, 2018, 47(9): 1238-1249.DOI: 10.11947/j.AGCS.2018.20170476. GU Yu, XU Ying, GUO Baofeng. Hyperspectral image classification by combination of spatial-spectral features and ensemble extreme learning machines[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(9): 1238-1249.DOI: 10.11947/j.AGCS.2018.20170476.
[5] 张兵. 高光谱图像处理与信息提取前沿[J]. 遥感学报, 2016, 20(5):1062-1090. ZHANG Bing. Advancement of hyperspectral image processing and information extraction[J]. Journal of Remote Sensing, 2016, 20(5):1062-1090.
[6] ZHENG Xiangtao, YUAN Yuan, LU Xiaoqiang. Dimensionality reduction by spatial-spectral preservation in selected bands[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(9): 5185-5197.
[7] HANG Renlong, LIU Qingshan. Dimensionality reduction of hyperspectral image using spatial regularized local graph discriminant embedding[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(9): 3262-3271.
[8] SUN Weiwei, DU Qian. Hyperspectral band selection: a review[J]. IEEE Geoscience and Remote Sensing Magazine, 2019, 7(2): 118-139.
[9] 张红, 吴智伟, 王继成, 等. 高光谱图像分类的Wasserstein配置熵非监督波段选择方法[J]. 测绘学报, 2021, 50(3): 405-415.DOI: 10.11947/j.AGCS.2021.20200006. ZHANG Hong, WU Zhiwei, WANG Jicheng, et al. 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.DOI: 10.11947/j.AGCS.2021.20200006.
[10] CHANG C I, DU Q, 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.
[11] CHANG C I, WANG S. Constrained band selection for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(6): 1575-1585.
[12] XU Buyun, LI Xihai, HOU Weijun, et al. A similarity-based ranking method for hyperspectral band selection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(11): 9585-9599.
[13] AHMAD M, HAQ I, MUSHTAQ Q, et al. A new statistical approach for band clustering and band selection using K-means clustering[J]. International Journal of Engineering Technology, 2011, 3(6): 606-614.
[14] 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.
[15] WANG Qi, ZHANG Fahong, LI Xuelong.Optimal clustering framework for hyperspectral band selection[EB/OL].[2023-02-10]. https://arxiv.org/abs/1904.13036.pdf.
[16] WANG Jun, TANG Chang, ZHENG Xiao, et al.Graph regularized spatial-spectral subspace clustering for hyperspectral band selection[J]. Neural Networks, 2022, 153: 292-302.
[17] ZHANG Aizhu, MA Ping, LIU Sihan, et al. Hyperspectral band selection using crossover-based gravitational search algorithm[J]. IET Image Processing, 2019, 13(2): 280-286.
[18] GENG Xiurui, SUN Kang, JI Luyan, et al. A fast volume-gradient-based band selection method for hyperspectral image[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(11): 7111-7119.
[19] WANG Qi, ZHANG Fahong, LI Xuelong. Hyperspectral band selection via optimal neighborhood reconstruction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(12): 8465-8476.
[20] WANG Qi, LI Qiang, LI Xuelong. Hyperspectral band selection via adaptive subspace partition strategy[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(12): 4940-4950.
[21] SUN He, REN Jinchang, ZHAO Huimin, et al. Adaptive distance-based band hierarchy (ADBH) for effective hyperspectral band selection[J]. IEEE Transactions on Cybernetics, 2022, 52(1): 215-227.
[22] WANG Qi, LI Qiang, LI Xuelong. A fast neighborhood grouping method for hyperspectral band selection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(6): 5028-5039.
[23] CAI Yaoming, LIU Xiaobo, CAI Zhihua. BS-nets: an end-to-end framework for band selection of hyperspectral image[J]. IEEE Tran-sactions on Geoscience and Remote Sensing, 2020, 58(3): 1969-1984.
[24] TSCHANNERL J, REN Jinchang, ZABALZA J, et al. Segmented autoencoders for unsupervised embedded hyperspectral band selection[C]//Proceedings of the 7th European Workshop on Visual Information Processing. Tampere:IEEE, 2018: 1-6.
[25] RODRIGUEZ A, LAIO A. Clustering by fast search and find of density peaks[J]. Science, 2014, 344(6191): 1492-1496.
[26] SUN Kang, GENG Xiurui, JI Luyan. Exemplar component analysis: a fast band selection method for hyperspectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(5): 998-1002.
[27] JIA Sen, TANG Guihua, ZHU Jiasong, et al. A novel ranking-based clustering approach for hyperspectral band selection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(1): 88-102.
[28] LUO Xiaoyan, XUE Rui, YIN Jihao. Information-assisted density peak index for hyperspectral band selection[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(10): 1870-1874.
[29] LI Qiang, WANG Qi, LI Xuelong. An efficient clustering method for hyperspectral optimal band selection via shared nearest neighbor[J]. Remote Sensing, 2019, 11(3): 350.
[30] YANG Rongchao, KAN Jiangming. An unsupervised hyperspectral band selection method based on shared nearest neighbor and correlation analysis[J]. IEEE Access, 2019, 7: 185532-185542.
[31] GUO Zhishuai, HUANG Tianyi, CAI Zhiling, et al. A new local density for density peak clustering[C]//Proceedings of 2018 Pacific-Asia Conference on Knowledge Discovery and Data Mining.Cham: Springer,2018: 426-438.
[32] KORN F, MUTHUKRISHNAN S. Influence sets based on reverse nearest neighbor queries[J]. ACM Sigmod Record, 2000, 29(2): 201-212.
[33] JARVIS R A, PATRICK E A. Clustering using a similarity measure based on shared near neighbors[J]. IEEE Transactions on Computers, 1973, 100(11): 1025-1034.
[34] ERTOZ L, STEINBACH M, KUMAR V. A new shared nearest neighbor clustering algorithm and its applications[C]//Proceedings of 2002 Workshop on Clustering High Dimensional Data and Its Applications at 2nd SIAM International Conference on Data Mining.[S.l.]:IEEE,2002.
[35] VAZQUEZ-FERNANDEZ E, DACAL-NIETO A, MARTIN F, et al. Entropy of gabor filtering for image quality assessment[C]//Proceedings of 2010 International Conference Image Analysis and Recognition.Berlin: Springer,2010: 52-61.
[36] 赵海士, 路来君, 杨晨. 一种基于图像熵的密度峰值聚类波段选择方法[J]. 吉林大学学报(理学版), 2017, 55(2): 376-378. ZHAO Haishi, LU Laijun, YANG Chen. A method for density peaks clustering band selection based on image entropy[J]. Journal of Jilin University (Science Edition), 2017, 55(2): 376-378.