Abstract: Vein based multi-modal biometric recognition is becoming an emerging need in order to have better security. Here, we have utilized dorsal and finger vein biometrics present in the hand. After preprocessing, the necessary features are extracted from dorsal vein imagesusing automatic thresholding and from finger vein images using hyper analytic wavelet transform. Once the vein features are extracted, vein patterns are generated based on the vein location. The vein pattern is large in length, so the selection of blocks from the large length vein patterns is important without affecting the accuracy. Here, firefly algorithm is utilized to select the optimal blocks. After selecting the optimal blocks, the location and size of blocks are fixed. Using these blocks matching is done with the block features stored in the database. The experimentation is carried out using the standard data bases of finger and dorsal hand vein images. The performance of the proposed technique is evaluated using the false acceptance and false rejection rates. The experimental results show that the proposed method achieved improved performance compared to the existing techniques.
S. Bharathi and R. Sudhakar, 2016. Optimal Block Selection for Vein Based Multimodal Biometrics. Asian Journal of Information Technology, 15: 4363-4369.