Key points are not available for this paper at this time.
Hyperspectral data analysis has been of importance to the remote sensing community, and analysis tools for extracting information (targets) from hyperspectral data have been developed in recent years. However, due to the vast amount of data available from hyperspectral signatures, it is important to develop fast and reliable methods for extracting useful information. In this paper, a combined derivative spectroscopy and Savitzky-Golay filtering method for the analysis of hyperspectral data is presented. This method is based on the concept of smoothing reflectance spectra in order to eliminate instrumental noise and then extracting absorption band positions (wavelengths) using high-order derivatives. Results from hyperspectral signatures for cotton, sicklepod, and bare soil are presented and a statistical analysis comparison to known absorption characteristics in terms of an error measure is performed to illustrate the applicability and validity of this method.
Ruffin et al. (Tue,) studied this question.