[OPTICAL REVIEW Vol. 17, No. 4 (2010) 367-370]
© 2010 The Japan Society of Applied Physics

Band Selection of Hyperspectral-Image Based Weighted Indipendent Component Analysis

Mojtaba Amini OMAM and Farah TORKAMANI-AZAR*

Cognitive Telecommunications Research Group, Faculty of Electrical and Computer Engineering, Shahid Beheshti University, G.C. Evin, Tehran, Iran

(Received April 6, 2010; Accepted June 4, 2010)

Huge amounts of data in hyperspectral images have been caused to represent approaches for the band selection of these images. In this paper, a new approach based on independent component analysis (ICA) is proposed. The idea of projection pursuit is used to order the bands on the basis of a non-gaussianity distribution. Applying a negentropy function to weight bands is a novel idea that leads to the selection of bands with minimum mutual information (MI) and besides maximum entropy, with respected to the bands selected using other methods.

Key words: hyperspectral image, band selection, independent components analysis, projection pursuit (PP)

*E-mail address: f-torkamani@sbu.ac.ir