A biometric authentication method is proposed using hyperspectral datacube of palm. The datacube is acquired by a recently developed equipment which captures narrow spectral band reflectances across a contiguous portion of the 396.37nm-990.64nm range with 0.93 nm spectral resolution in approximately 10 seconds. The datacube gives spectrally resolved images with narrow spectral bands. The acquired reflectance value datacube represents the structures associated with the distributions of various biological substances.
The proposed method first computes spatial correlation between a test data and a set of stored template data at each wavelength of the spectrally resolved image. Then the algorithm computes a score by integrating the spatial correlations with respect to wavelength. The method appears functional for biometric authentication even though the hardware was designed for melanoma detection.