Band selection from statistical wavelet models

The high data dimensionality of hyperspectral images increases the burden on data computation, storage, and transmission; fortunately, the high redundancy in the spectral domain allows for significant dimensionality reduction. Band selection provides a simple dimensionality reduction scheme by discarding bands that are highly redundant, therefore preserving the structure of the dataset. This paper proposes a new criterion for pointwise ranking-based band selection that uses a non-homogeneous hidden Markov chain (NHMC) model for redundant wavelet coefficients of the hyperspectral signature. The model provides a binary-valued multiscale label that encodes the presence of discriminating spectral information at different bands. A band ranking score considers the average correlation among the class-wise average NHMC labels for each band. Wavelet-based features provide robustness to noise thanks to the multiscale analysis performed by the transform. We also test richer label vectors that provide a more finely grained quantization of spectral fluctiations. In addition, since band selection methods based on band ranking often ignore correlations in selected bands, we include an optional redundancy elimination step and test its effect on band selection performance. Experimental results include a comparison with several relevant supervised band selection techniques.