Fusion of Multi-spectral Palmprint Images for ...

Biometric system has been actively emerging in various industries for the past few years and it is continuing to roll to provide higher security features for access control system. In this paper, we propose a multi-modal system for person identification using multi-spectral palmprint images. In this work, the observation vectors are extracted using 2D Gabor filters. Thus, the images {RED, BLUE, GREEN and Nearest-Infrared (NIR)} are filtered by a 2D Gabor filter with different orientations and then compressed using the PCA. Subsequently, we use the GMM and HMM for modeling the observation vector of each band. Finally, log-likelihood scores are used for palmprint matching. The multi-modal systems fuse information from several modalities in order to achieve better identification performance. Therefore, all bands are integrated using the fusion at the matching score level in order to construct an efficient multi-modal identification system. The experimental results showed that the designed system achieves an excellent identification rate and provide more security than uni-modal system.