Dorsal Hand Vein Biometric Recognition Based on Orientation of Local Binary Pattern
K Nadiya, Varun P. Gopi · 2020
As the security system in society increases, vein recognition has become an authentic method for biometrics-based identification systems. Biometrics is a field of statistical analysis of the biological characteristics of an individual. Here, the dorsal hand vein is used for biometric identification. The dorsal hand vein biometric system is one of the emerging biometric technologies to determine or authenticate the identity of an individual. In the proposed method, the input dorsal hand vein image is first preprocessed to remove unwanted noises. The feature, which is directional information about vein, is extracted from the preprocessed image using an extended version of Local Binary Pattern (LBP) called the orientation of LBP (OLBP). The feature extracted using OLBP is represented in binary format so that the matching of features can be carried out by a Hamming distance. The proposed method tested with a standard database from the University of Cairo's Department of Biomedical and Systems Engineering. The results show that the suggested method gives good performance with 98.4% of Correct Recognition Rate, 0.07 of False Rejection Rate, and 0.08 of False Acceptance Rate and it is suitable for the biometric recognition system. An efficient method for dorsal hand vein biometric recognition is developed using OLBP. The comparison with other methods produced results that reveal the supremacy of the proposed algorithm for biometric recognition.