Visible spectrum bands of palmprint image for a robust biometric identification system

Biometric is known a large attention in the recent years. In the most biometric based security systems, the personal identity recognition depend on their behavioral, biological or physical characteristics. Among several biometric technologies, palmprint technology has a wide spread in the biometric systems for a several advantages such as the user-friendly and the low cost. In this paper, a palmprint identification system is proposed. The palmprint image is captured under several wave lights (several bands), then a clustering process is used in order to reduce the number of bands. However, Gabor filter technique is used to extract the palmprint features. The proposed system is tested and evaluated using hyper-spectral palmprint database of 300 persons. The obtained results show that the designed system achieves an excellent identification rate.