Robust linear spectral unmixing using outlier detection

Abstract

This paper presents a Bayesian algorithm for linear spectral unmixing that accounts for outliers present in the data. The proposed model assumes that the pixel reflectances are linear mixtures of unknown endmembers, corrupted by an additional term modelling outliers and additive Gaussian noise. A Markov random field is considered for outlier detection based on the spatial and spectral structures of the anomalies. This allows outliers to be identified in particular regions and wavelengths of the data cube. A Bayesian algorithm is proposed to estimate the parameters involved in the model yielding a joint linear unmixing and outlier detection algorithm. Simulations conducted with synthetic data demonstrate the accuracy of the proposed unmixing and outlier detection strategy for the analysis of hyperspectral images.