Novel Algorithm for VLOB IRIS code Database Organization and Adaptive Searching for IRIS code Similarity

The Iris recognition has become as one of the promising biometrics feature in human identification system. In addition to conventional IRIS capture challenges, IRIS code searching and similarity matching of IRIS code has also become a challenging task in IRIS code VLOB (Very Large Object database). Hamming Distance (HD), a sequential similarity matching technique, proves in-efficient for VLOB Iris code database objects. With technology growth the bit granularity has improved. This has resulted in more accuracy due to higher bit density. This high granularity leads to variable size IRIS codes rather than fixed sized IRIS codes hence adaptively is one major challenge. This paper proposes a novel, adaptive IRIS code database organization and searching algorithm proposing the improvement in template similarity matching process. The algorithm reduces the redundant IRIS code comparisons resulting in faster searching speed. For the IRIS that has accumulated the cataract, HD methods shows major failure results. The proposed IRIS code organization algorithm shows better performance over conventional HD method of template matching and IRIS code selection for template similarity matching. KeywordsDatabase Organization, Iris codes Descriptors, Iris Code Searching , and Iris Code Similarity.