Adaptive Threshold for Fingerprint Recognition System Based on Threat Level and System Load
Fingerprint is widely used trait for person recognition in civilian applications. A user is authenticated when matching score is greater than acceptance threshold. The performance of fingerprint system (FS) is evaluated based on false acceptance rate (FAR) and false rejection rate (FRR). Usually the FS is set to work at a rate where FAR and FRR are equal (EER). However, operating at EER allows finite FAR which is risky during critical threat. In response acceptance threshold must shifts towards zero FAR to mitigate threat. This increases FRR, system load and user inconvenience. In civilian application acceptance threshold is set by vendor and currently there is no research attempt to change it dynamically. This is necessary as; 1) system must respond to external parameters like load and threat level; 2) system must balance security and user convenience due to high traffic?c. This paper describes a method to change acceptance threshold over the interval EER to zero FAR based on system load and threat level. The proposed method is based on fuzzy inference system (FIS) and artificial neural network (ANN).