Abstract
Contiguous narrow bands of hyperspectral images greatly increase computational complexity. Redundancy reduction is therefore necessary. Here, a minimum redundancy and maximum variance based unsupervised band selection methodology is proposed. Discrete wavelet transformation is applied on the data to reduce spatial redundancy without much effecting the overall band correlations. This in turn made the process more time efficient and noise resilient. Highly correlated bands are considered similar, and one with higher variance is accepted as being more discriminating. Finally, classification is performed with the selected bands and overall accuracy (OA) is calculated. The proposed method is compared with four other existing state-of-the-art methods in the similar field in terms of OA and execution time for evaluating the performance.
Similar content being viewed by others
References
An D et al (2016) Hyperspectral field estimation and remote-sensing inversion of salt content in coastal saline soils of the Yellow River Delta. Int J Remote Sens 37(2):455–470
Deepa P, Thilagavathi K (2015) Feature extraction of hyperspectral image using principal component analysis and folded-principal component analysis. In: 2nd International conference on electronics and communication systems (ICECS), IEEE, pp 656–660
Devijver PA, Kittler J (1982) Pattern recognition: a statistical approach, 1st edn. Prentice-Hall International, New Delhi
Ding L, Tang P, Li H (2013) Isomap-based subspace analysis for the classification of hyperspectral data. In: IEEE international geoscience and remote sensing symposium (IGARSS). IEEE, pp 429–432
Duda RO, Hart PE, Stork DG (2000) Pattern classification. Wiley, New York
Feng J, Jiao L, Liu F, Sun T, Zhang X (2016) Unsupervised feature selection based on maximum information and minimum redundancy for hyperspectral images. Pattern Recogn 51:295–309
Feng J, Jiao L, Sun T, Liu H, Zhang X (2016) Multiple kernel learning based on discriminative kernel clustering for hyperspectral band selection. IEEE Trans Geosci Remote Sens 54(11):6516–6530
Fukunaga K (2013) Introduction to statistical pattern recognition. Elsevier, Amsterdam
Ghorbanian A, Mohammadzadeh A (2018) An unsupervised feature extraction method based on band correlation clustering for hyperspectral image classification using limited training samples. Remote Sens Lett 9(10):982–991. https://doi.org/10.1080/2150704X.2018.1500723
Gonzalez RC, Woods R (2007) Digital image processing. Pearson Prentice Hall, Upper Saddle River
Götze C, Beyer F, Gläßer C (2016) Pioneer vegetation as an indicator of the geochemical parameters in abandoned mine sites using hyperspectral airborne data. Environ Earth Sci 75:613
Hughes G (1968) On the mean accuracy of statistical pattern recognizers. IEEE Trans Inf Theory 14(1):55–63. https://doi.org/10.1109/TIT.1968.1054102
Jain AK, Duin RPW, Mao J (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal Mach Intell 22(1):4–37
Jia S, Tang G, Zhu J, Li Q (2016) A novel ranking-based clustering approach for hyperspectral band selection. IEEE Trans Geosci Remote Sens 54(1):88–102
Kim B, Landgrebe DA (1991) Hierarchical classifier design in highdimensional numerous class cases. IEEE Trans Geosci Remote Sens 29(4):518–528
Launeau P et al (2017) Airborne hyperspectral mapping of trees in an urban area. Int J Remote Sens 38(5):1277–1311
Li W, Prasad S, Fowler JE, Bruce LM (2012) Locality-preserving dimensionality reduction and classification for hyperspectral image analysis. IEEE Trans Geosci Remote Sens 50(4):1185–1198
Paul A, Bhattacharya S, Dutta D, Sharma JR, Dadhwal VK (2015) Band selection in hyperspectral imagery using spatial cluster mean and genetic algorithms. GISci Remote Sens 52(6):644–661. https://doi.org/10.1080/15481603.2015.1075180
Sugiyama M (2007) Dimensionality reduction of multimodal labeled data by local fisher discriminant analysis. J Mach Learn Res 8:1027–1061
Sun K, Geng X, Ji L (2015) Exemplar component analysis: a fast band selection method for hyperspectral imagery. IEEE Geosci Remote Sens Lett 12(5):998–1002
Sun K, Geng X, Ji L, Lu X (2014) A new band selection method for hyperspectral image based on data quality. IEEE J Sel Top Appl Earth Observ Remote Sens 7(6):2697–2703
Sun Y, Wang S, Liu Q, Hang R, Liu G (2017) Hypergraph embedding for spatial-spectral joint feature extraction in hyperspectral images. Remote Sens 9(5):506. https://doi.org/10.3390/rs9050506
Upadhyay V, Kumar A (2018) Hyperspectral remote sensing of forests: technological advancements, opportunities and challenges. Earth Sci Inf 11:487. https://doi.org/10.1007/s12145-018-0345-7
Wang C, Gong M, Zhang M, Chan Y (2015) Unsupervised hyperspectral image band selection via column subset selection. IEEE Geosci Remote Sens Lett 12(7):1411–1415
Wang Q, Lin J, Yuan Y (2016) Salient band selection for hyperspectral image classification via manifold ranking. IEEE Trans Neural Netw Learn Syst 27(6):1279–1289
Yang H, Du Q, Su H, Sheng Y (2011) An efficient method for supervised hyperspectral band selection. IEEE Geosci Remote Sens Lett 8(1):138–142
Yuan H, Tang YY (2015) Learning with hypergraph for hyperspectral image feature extraction. Geosci Remote Sens Lett IEEE 12(8):1695–1699
Yuan Y, Lin J, Wang Q (2016) Dual-clustering-based hyperspectral band selection by contextual analysis. IEEE Trans Geosci Remote Sens 54(3):1431–1445
Zarco-Tejada P, González-Dugo M, Fereres E (2016) Seasonal stability of chlorophyll fluorescence quantified from airborne hyperspectral imagery as an indicator of net photosynthesis in the context of precision agriculture. Remote Sens Environ 179:89–103
Zhang W, Li X, Zhao L (2018) A fast hyperspectral feature selection method based on band correlation analysis. IEEE Geosci Remote Sens Lett 15(11):1750–1754
Zhao W, Du S (2016) Spectral–spatial feature extraction for hyperspectral image classification: a dimension reduction and deep learning approach. IEEE Trans Geosci Remote Sens 54(8):4544–4554. https://doi.org/10.1109/TGRS.2016.2543748
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
Paul, A., Chaki, N. Dimensionality Reduction Using Band Correlation and Variance Measure from Discrete Wavelet Transformed Hyperspectral Imagery. Ann. Data. Sci. 8, 261–274 (2021). https://doi.org/10.1007/s40745-019-00210-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40745-019-00210-x