Coarse indexing of iris database based on iris colour
Niladri Bihari Puhan, N. Sudha · International Journal of Biometrics · 2011
This paper presents a new iris indexing technique based on iris colour. Blue and red indices are computed from the chrominance values of the pixels and indexing methods are proposed by combining these indices. The performance measures such as the hit rate and the penetration rate computed on the colour iris databases, namely UBIRIS (Proenca and Alexandre, 2005), and UPOL (Dobeš and Machala, 2004), show the effectiveness of the iris colour for indexing large iris databases. From the experimental results, it is observed that the indexing method with set intersection-based combination of indices achieves the best performance with a hit rate above 98% and the penetration rate less than 25%.