Indexed on: 12 Jan '16Published on: 12 Jan '16Published in: Computer Science - Distributed; Parallel; and Cluster Computing
The exponential growth of enrollment in the biometric system produces massive amount of high dimensionality data that leads to the degradation of performance of the system. Also different kind of issues pops up in handling such huge data sets. Therefore in order to overcome the performance issues arising due to this data deluge we aim to implement this system in a distributed manner on a high performance cluster cloud. High availability and scalability are some of the added advantages of using this approach. In this paper we propose a mobile biometric system that uses dynamic signatures and performs authentication on the go. This framework includes steps involving data capture using any handheld mobile device, then storage, preprocessing and training in a distributed manner overcloud. Moreover, the methodology adopted by us is very novel as it provides better time efficiency and scalability. We have systematically designed and conducted experiments to evaluate the proposed framework on dynamic signature datasets recorded in real time. The experiments demonstrate that better performance was achieved by our proposed framework as compared to the other methods used in recent literature.