Is spectral reflectance of the face a reliable biometric?

摘要: Over a decade ago, Pan et al. [IEEE TPAMI 25, 1552 (2003)] performed face recognition using only the spectral reflectance of at six points and reported around 95% rate. Since their database is private, no one has been able to replicate these results. Moreover, due unavailability public datasets, there detailed study in literature on viability facial for person identification. In this study, we introduce new profiles measured with high precision spectrometer. For each 40 subjects, was same as multiple sessions time lapse. Furthermore, sample from two hyperspectral image datasets analyzed data state art classification techniques. The best performing classifier achieved maximum rank-1 identification rate 53.8%. We conclude that alone not reliable biometric unconstrained recognition.