Band Selection Oriented to Easy-Confused Objects for Classification of Hyperspectral Imagery

Abstract:

As a dimensionality reduction technique, band selection is an importance pre-processing step for classifiers. In this paper, a band selection approach oriented to easy-confused objects for classification of hyper spectral imagery is presented. Firstly, an Objects Confusion Index (OCI) is established to ascertain the easy-confused objects. Then the two band selection schemes, that are two-class mode and multi-class mode, are designed by adopting Bhattacharyya distance as class reparability measure.

• Periodical: Key Engineering Materials (Volume 500)
• Pages: 355-361
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.500.355
• Online since: January 2012
• Authors: Xin Zhao, Xing Li
• Keywords: Band Selection, Classification, Easy-Confused Objects, Hyper-Spectral

References

[1] Hughes G. On the mean accuracy of statistical pattern recognizers. Ieee Transactions on Information Theory, 1968, 14(1): 55-63.

DOI: 10.1109/tit.1968.1054102

[2] Molina LC, Belanche L, Nebot A. Feature selection algorithms: a survey and experimental evaluation. Proceedings 2002 IEEE International Conference on Data Mining. ICDM 2002. IEEE Comput. Soc, Los Alamitos, CA, USA, 306-313.

DOI: 10.1109/icdm.2002.1183917

[3] Huang R, He MY. Band selection based on feature weighting for classification of hyperspectral data. Ieee Geoscience and Remote Sensing Letters, 2005, 2(2): 156-159.

DOI: 10.1109/lgrs.2005.844658

[4] Guo BF, Damper RI, Gunn SR, et al. A fast separability-based feature-selection method for high-dimensional remotely sensed image classification. Pattern Recognition, 2008, 41(5): 1653-1662.

DOI: 10.1016/j.patcog.2007.11.007

[5] Raudys S, Pikelis V. On dimensionality, sample size, classification error, and complexity of classification algorithm in pattern recognition. Ieee Transactions on Pattern Analysis and Machine Intelligence, 1980, PAMI-2(3): 242-252.

DOI: 10.1109/tpami.1980.4767011

[6] Xia J-T, He M-Y. Feature extraction in kernel space using Bhattacharyya distance as criterion function. Chinese Journal of Computers, 2004, 27(5): 683-689.

[7] Miao X, Gong P, Swope S, et al. Detection of yellow starthistle through band selection and feature extraction from hyperspectral imagery. Photogrammetric Engineering and Remote Sensing, 2007, 73(9): 1005-1015.

[8] Reyes-Aldasoro CC, Bhalerao A. The Bhattacharyya space for feature selection and its application to texture segmentation. Pattern Recognition, 2006, 39(5): 812-826.

DOI: 10.1016/j.patcog.2005.12.003