Hand based biometric system and its performance analysis in person authentication

Abstract

Biometric system is the advancing trend in the technical era of identifying the individual, and it is mainly employed for various applications, especially in the field of security, banking, immigration, forensic labs, and so on. Biometric system used for identifying the person is a necessary as the traditional methods have utilized the handwriting and the signature of the person, for the purpose of person identification. Thus, the biometric system emerges as the trending topic in the field of research, by using the hand-based images to identify the person. The main aim of this research is to identify and authenticate the person based on the hand-based features, such as palmprint, finger vein, and finger knuckle images. The purpose of extracting features from these images is that these images hold rich contents of information of the person. The present research concentrates on the multimodal biometric recognition system that uses two images for feature extraction rather than the unimodal biometric system. The features extracted from the images are fused together using a weighing score and the value of the weighing score is determined optimally by using the newlineFractional Fire Fly (FFF) optimization algorithm and the weight score is determined using the proposed Fractional Cuckoo Search (FCS) optimization algorithm. The fused features represent the feature vector, which is fed as the input to the proposed classifier for the recognition. newline newline