Abstract
In order to avoid the problem of over-dependent on the similarity to select the band but ignoring the amount of information and the correlation of the band, this paper proposes a method of hyperspectral image feature perception based on clustering and intra-class frequency band index. In this paper, the band texture feature vectors are used to calculate the similarity matrix. Then affine propagation clustering algorithm clusters all the bands according to the similarity matrix. The band with the largest intra-class band index in a certain cluster is selected as the representative band, so as to achieve the purpose of band selection. Finally, support vector machine classification algorithm is used to classify the objects on the image after band selection. By combining the Affine Propagation algorithm and Intra-Class Band Index, this paper proposed the AP-ICBI algorithm so that the band with large amount of information and small correlation can be selected in the high-quality band clustering results. In the experiment, the overall classification accuracy (OA), the Kappa coefficient and user accuracy (UA) are taken as the evaluation indexes. The experimental results showed that the proposed AP-ICBI algorithm can effectively improve the classification accuracy comparing with other methods.
Similar content being viewed by others
References
Chavez PS, Berlin GL, Sowers LB (1982) Statistical method for selecting Landsat-MSS ratios. J Appl Photogr Eng 8(1):23–30
Datta A, Ghosh S, Ghosh A (2017) Unsupervised band extraction for hyperspectral images using clustering and kernel principal component analysis. Int J Remote Sens 38(3):850–873
Frey BJ (2007) Clustering by passing messages between data points. Science 315(5814):972–976
Haitao L, Haiyan G, Bing Z (2007) Research on hyperspectral remote sensing image classification based on MNF and SVM. Remote Sens Inf 5:12–15
Hosseini SA (2011) A new fast algorithm for multiclass hyperspectral image classification with SVM. Int J Remote Sens 32(23):8657–8683
Li C, Chu H, Kuo B et al (2011) Hyperspectral image classification using spectral and spatial information based linear discriminant analysis. IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp 1716–1719
Li X, Zhao C, Wang Y (2015) Sparse representation within disconnected spatial support for target detection in hyperspectral imagery. International Conference on Signal Processing Proceedings, ICSP. 2015. 802-806. https://doi.org/10.1109/ICOSP.2014.7015114
Luo L, Wang X, J N (2012) Remote sensing forest classification with text based on ICA and SVM. Comput Eng Appl 48(13):227–229
Majdar RS, Ghassemian H (2017) A probabilistic SVM approach for hyperspectral image classification using spectral and texture features. Int J Remote Sens 38(15):4265–4284
Martínez-Usó A, Pla F, Sotoca JM et al (2007) Clustering-based hyperspectral band selection using information measures. IEEE Trans Geosci Remote Sens 45(12):4158–4171
Nakamura RYM, Fonseca LMG (2014) Nature-inspired framework for hyperspectral band selection. IEEE Trans Geosci Remote Sens 52(4):2126–2137
Qian Y, Yao F, Jia S (2009) Band selection for hyperspectral imagery using affinity propagation. IET Comput Vis 3(4):213–222
Qin F, Zhang A (2015) Hyperspectral band selection based on spectral clustering and separability factor. Spectrosc Spectr Anal 35(5):1357–1364
Tian Y (2008) Research on dimensional reduction of hyperspectral remote sensing images. Harbin Engineering University, Harbin
Ulaby FT, Kouyate F, Brisco B (1986) Texttural information in SAR images. IEEE Trans Geosci Remote Sens 24(2):235–245
Wang Q, Zhang F, Li X (2020) Hyperspectral band selection via optimal neighborhood reconstruction. IEEE Trans Geosci Remote Sens. https://doi.org/10.1109/TGRS.2020.2987955.
Wang C, Menenti M, Li Z (2003) Modified principal component analysis (MPCA) for feature selection of hyperspectral imagery. In: Proceedings of the IEEE Conference Geoscience and Remote Sensing Symposium (IGARSS) 6:3781–3783
Wang Y, Wu G, L D (2014) Plant species identification based on independent component analysis for hyperspectral data. J Softw 9(6):1532–1537
Wang Q, Yang G, Xiang Y (2017) Band selection method of hyperspectral image based on subspace partition. Ship Electron Eng 37(43):98–102
Wang Q, Li Q, Li X (2020) A fast neighborhood grouping method for hyperspectral band selection. IEEE Trans Geosci Remote Sens 58:8465–8476. https://doi.org/10.1109/TGRS.2020.3011002
Wei X, Cai L, Liao B, Lu T (2020) Local-view-assisted discriminative band selection with hypergraph autolearning for hyperspectral image classification. IEEE Trans Geosci Remote Sens 58(3):2042–2055
Yang J, Wang L, J Q (2016) A new residual fusion classification method for hyperspectral images. Int J Remote Sens 37(4):745–769
Zhai H, Zhang H, Zhang L (2015) Spectral-spatial clustering of hyperspectral remote sensing image with sparse subspace clustering model. Hyperspectral Image Signal Proc Evol Remote Sens:1–4
Zhai H, Zhang H, Zhang L (2016) Squaring weighted low-rank subspace clustering for hyperspectral image band selection. IEEE Trans Geosci Remote Sens:2434–2237
Zhao C, Li X, Ren J, Marshall S (2013) Improved sparse representation using adaptive spatial support for effective target detection in hyperspectral imagery. Int J Remote Sens 34:8669–8684
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yan, Y., Yu, W. & Zhang, L. A method of band selection of remote sensing image based on clustering and intra-class index. Multimed Tools Appl 81, 22111–22128 (2022). https://doi.org/10.1007/s11042-021-11865-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-021-11865-1