Wavelet-based Analysis of Hyperspectral Data for Detecting Spectral Features

The purpose of feature extraction is to abstract substantial information from the original data input and filtering out redundant information. In this paper we transfer the hyperspectral data from the original-feature space to a scale-space plane by using a wavelet transform to extract significant spectral features. The wavelet transform can focus on localized signal structures with a zooming procedure. The absorption bands are thus detected with the wavelet transform modulus maxima, and the Lipschitz exponents, are estimated at each singularities point of the spectral curve from the decay of the wavelet transform amplitude. The local frequency variances provide some useful information about the oscillations of the hyperspectral curve for each pixel. Different type of materials can be distinguished on the basis of the differences in the local frequency variation. The new method generates more meaningful features and is more stable than other known methods for spectral feature extraction.