Information-theoretic noisy band detection in hyperspectral imagery

Hyperspectral imagery consists of hundreds of successive bands that carry spectral information about the underlying materials at various wavelengths. However, due to practical factors such as atmospheric effects and sensor characteristics, some spectral bands contain high amounts of noise. In this paper, an effective information-theoretic algorithm based on mutual information that automatically detects such noisy bands is proposed. The effectiveness and accuracy of the proposed approach is validated on hyperspectral images collected by the AVIRIS and TELOPS sensors. Experimental results show that the proposed method outperforms the other algorithms in the literature.