This paper describes the application of Independent Component Analysis (ICA) and Terahertz technologies for materials identification. Similar to Principal Component Analysis (PCA), Independent component analysis is a method for finding underlying factors or components from multivariate statistical data. In materials identification problems, the goal is to separate and identify materials from mixtures. We introduce a novel pseudo-inverse ICA-based filtering algorithm to remove noise from images or signals. Both spectroscopic and chromatographic techniques can be used for materials identification. In this paper, continuous-wave Terahertz technology is used to illuminate mixtures of materials. And Independent Component Analysis is used to help separate and identify materials based on the Terahertz images. In the examples given, explosives under different covers are separated and identified using this method.