Non-Negative Matrix Factorization with Orthogonality Constraints for Chemical Agent Detection in Raman Spectra

Abstract

We introduce non-negative matrix factorization with orthogonality constraints (NMF-OC) for detection of a target spectrum in a given set of Raman spectra data. An orthogonality measure is defined and two different orthogonality constraints are imposed on the standard NMF to incorporate prior information into the estimation and hence to facilitate the subsequent detection procedure. Experimental results are presented to compare NMF-OC with the basic NMF and ICA methods in detection, and to demonstrate its effectiveness in the chemical agent detection problem.