A Comparison Between Linear and Non-Linear Spectral Unmixing Methods

Abstract

Spectral unmixing is a key process in identifying the spectral signature of materials and quantifying their spatial distribution over an image. This paper aims to investigate linear and nonlinear methods used to solve spectral unmixing problems, the methods were compared based on their prediction accuracy, robustness against noise and computational time using laboratory simulated data. Results show that the nonlinear methods outperform the linear methods in terms of prediction accuracy and robustness against noise but are computationally more expensive.