Principal component analysis for compression of hyperspectral images

In this paper, we explore the possibility to use the principal component analysis for compression of hyperspectral images. When the principal component analysis is applied to AVIRIS data that has 220 channels, we found that most energy is concentrated on a few eigenvalues, indicating that it may be possible to compress hyperspectral images significantly. The performance of the proposed algorithm is evaluated in terms of SNR and classification accuracies of selected classes. Experiments with AVIRIS data show promising results.