Abstract
To solve the problem of low detection efficiency of present hyperspectral band selection methods, a new band selection method based on curve area and genetic theory (CAGT) is proposed in this paper. Primarily, the method uses the area under the Receiver Operating Characteristic (ROC) curve as a band selection criterion to measure the target detection effect of the band; then construct fitness function based on this criterion and use genetic algorithm to optimize the band selection. Finally, the band subset with better target detection result can be obtained. Therefore, both data dimensionality reduction and improvement of target detection result can be realized at the same time. Experimental results on a real world hyperspectral data show the efficiency of the proposed CAGT method to improve the detection performance.
Similar content being viewed by others
References
G.F. Hughes, On the mean accuracy of statistical pattern recognizers. IEEE Trans. Inf. Theory 14(1), 55–63 (1968)
S. Charles, Selecting band combination from multi- spectra1 data. Photogramm. Eng. Remote. Sens. 51(6), 681–687 (1985)
P.S. Chacvez, G.L. Berlin, L.B. Sowers, Statistical method for selecting Landsat MSS ratios. J. Appl. Photogr. Eng. 1(8), 23–30 (1982)
H. J. Su, P. J. Du, Y. H. Sheng, Study on band selection algorithms of hyperspectral image data. Application Research of Computers, 25(4): 1093–1096 (2008)
F. Liu, J.Y. Gong, A classification method for high spatial resolution remotely sensed image based on multi-feature. Geogr. and Geo-Inf. Sci. 25(3), 19–41 (2009)
Y. Li, A new bands selection algorithm for hyperspectral image using hyperspectral derivative on Clifford manifold. Inf. Technol. J. 11(7), 904–909 (2012)
M. Diani, N. Acito, M. Greco, G. Corsini, A new band selection strategy for target detction in hyperspectral images. Proc. 12th Int. Conf. Knowl-Based Intell. Inf. Eng. Syst. 3, 424–431 (2008)
R. Li, J. Liu et al., A systematic approach toward detection of seagrass patches from hyperspectral imagery. Mar. Geod. 35, 271–286 (2012)
H. Gholizadeh et al., A decision fusion framework for hyperspectral subpixel target detection. Photogrammetrie • Fernerkundung • Geoinformation 3, 267–280 (2012)
Acknowledgments
This work was supported by the National Natural Science Foundation of China under project No.41174093.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wang, Y., Huang, S., Liu, D. et al. A novel band selection method based on curve area and genetic theory. J Opt 43, 193–202 (2014). https://doi.org/10.1007/s12596-014-0199-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12596-014-0199-4