Evaluating the Potential of Clonal Selection Optimization Algorithm to Hyperspectral Image Feature Selection

Abstract:

Article Preview

The great number of captured near spectral bands in hyperspectral images causes the curse of dimensionality problem and results in low classification accuracy. The feature selection algorithms try to overcome this problem by limiting the input space dimensions of classification for hyperspectral images. In this paper, immune clonal selection optimization algorithm is used for feature selection. Also one of the fastest Artificial Immune classification algorithms is used to compute fitness function of the feature selection. The comparison of the feature selection results with genetic algorithm shows the clonal selection’s higher performance to solve selection of features.

Info:

Periodical:

Edited by:

David Wang

Pages:

799-805

DOI:

10.4028/www.scientific.net/KEM.500.799

Citation:

F. Samadzadegan et al., "Evaluating the Potential of Clonal Selection Optimization Algorithm to Hyperspectral Image Feature Selection", Key Engineering Materials, Vol. 500, pp. 799-805, 2012

Online since:

January 2012

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.