Applying fusion in thermal face recognition

Abstract

Face recognition based on thermal images has minor importance in comparison to visible light spectrum recognition. Nevertheless, in applications such as livelyness detection or fever scan, thermal face recognition is used as a stand-alone module, or as part of a multi-modal biometric system. This paper investigates combinations of many methods, used for thermal face recognition, and introduces some new and modified algorithms, which have not been used in the area as of yet. Moreover, we show that the best method is always limited to a certain database (input data). In order to address this problem, the multi-algorithmic biometric fusion, based on the logistic regression, is deployed.