Fingerprint Indexing Using Minutiae and Pore Features
Source: West Virginia University
In this paper, the authors propose level-2 and level-3 feature based fingerprint indexing algorithm to improve the speed and accuracy of identification. Indexing parameters are computed using the minutiae and pore features. The identification performance is further improved by incorporating Dempster Shafer theory based match score fusion algorithm. Experimental results on a high resolution finger-print database show that the proposed algorithm improves the identification performance by at least 10% compared to existing fingerprint identification algorithms.