Novel Multiple Impression based Multimodal Fingerprint Recognition System

As population grow, the task of fingerprint verification system becomes increasingly more difficult. However performance improvement of biometric based verification system is an interesting and emerging area. In fingerprint verification systems, there are usually multiple enrolled impressions for a same finger. The performance of the systems can be improved by combining these impressions at enrollment level as well as testing level. In this research paper we use a method of score level fusion using multiple enrollment and multiple testing impressions to achieve higher accuracy. Experimental results show that a larger improvement can be obtained by using fusion of multiple enrollments and one testing impression scheme than other one to one matching. Experiment conducted with FVC2000, FVC 2002 and our own (BAMU) database. For the evaluation performance measures (FAR), (FRR), Recognition rate (RR) Equal Error Rate (EER) and Receiver Operating Characteristics curve (ROC) has been used.