Application of Multivariate Curve Resolution Alternating Least Squares (MCR-ALS) to remote sensing hyperspectral imaging

The application of the MCR-ALS method is demonstrated on two simulated remote sensing spectroscopic images and on one experimental reference remote sensing spectroscopic image obtained by the Airborn Visible/Infrared Imaging Spectrometer (AVIRIS). By application of MCR-ALS, the spectra signatures of the pure constituents present in the image and their concentration distribution at a pixel level are estimated. Results obtained by MCR-ALS are compared to those obtained by other methods frequently used in the remote sensing spectroscopic imaging field like VCA and MVSA. In the case of the analysis of the experimental data set, the resolved pure spectra signatures were compared to reference spectra from USGS library for their identification. In all cases, results were also evaluated for the presence of rotational ambiguities using the MCR-BANDS method. The obtained results confirmed that the MCR-ALS method can be successfully used for remote sensing hyperspectral image resolution purposes. However, the amount of rotation ambiguity still present in the solutions obtained by this and other resolution methods (like VCA or MVSA) can still be large and it should be evaluated with care, trying to reduce its effects by selecting the more appropriate constraints. Only in this way it is possible to increase the reliability of the solutions provided by these methods and decrease the uncertainties associated to their use.

This article summarizes the use of the MCR-ALS method as a powerful tool for the resolution of hyperspectral images on their constituents. Non-negativity, spectral normalization and local rank constraints are used to get physical meaningful resolution results and decrease rotation ambiguities. MCR-BANDS method is used to evaluate the presence and extend of rotational ambiguities.Figure optionsDownload as PowerPoint slideHighlights
► MCR-ALS is successfully applied to remote sensing hyperspectral images.
► Pure spectra and relative concentrations of image constituents were obtained.
► MCA-ALS results are favorably compared to results obtained by MVSA and VCA methods.
► Physical constraints were implemented to decrease the rotational ambiguities.
► MCR-BANDS is used to evaluate the presence and extent of rotational ambiguities.