A palmprint based recognition system for smartphone

This paper proposes a palmprint based human identification system for smartphone. It presents simple markers to the user for palmprint video acquisition and suggests a strategy to autimatically select the best frame for recognition. Performance of the system has been evaluated by using two popular feature extraction techniques viz. SIFT and ORB. Histogram equalization is used for image enhancement. A palmprint database of 51 users has been collected by using Samsung Galaxy Grand I-9082 smartphone to test the system. Experimental results have shown acceptable system performance by achieving correct recognition rate (CRR) and equal error rate (EER) of 96.07% and 5.5% respectively on the database.