An Effective Multimodal Biometric System Based on Textural Feature Descriptor

In recent years the upsurge of contemporary technical assets results in the augmented prerequisite of precise and robust security systems. The utmost prevailing alternative for the security requirements is the deployment of effective biometric authentication frameworks. The unimodal biometric frameworks have various snags due to which they lag in precision. In this concern, the multimodal biometrics has gained more consideration of investigators as it utilizes the proven and effective strategy to integrate the multiple human traits for authentication. This paper proposes a multimodal biometric framework, which is based upon score level fusion of iris, palmprint, and finger-knuckle-print (FKP). It is a touchless system as these traits can be acquired without any physical contact with the sensors, so this system maintains the hygiene factor, which is the need of the hour in this pandemic time. Rigorous experimentations on three benchmark databases reveal that the proposed multimodal framework is proficient in augmenting the recognition rate in comparison to the unimodal framework as well as prevailing multimodal state of art works on attaining lowest equal error rate (EER) of 0.56% and highest genuine acceptance rate (GAR) of 99.70%.