Biometric authentication of physical characteristics recognition using artificial neural network with PSO algorithm

Abstract

Biometric authentication is the verification and identification of a person uniquely based on physical characteristics such as fingerprint, palm print and knuckle print. Biometric authentication is used in computer science as a form of identification and access control. For authentication four different modules are operated, including pre-processing, feature extraction, fusion and recognition modules, which are performed in our proposed work. In pre-processing module various techniques are used to improve the image quality and render the image suitable for additional processing; then for each character different feature extraction techniques are attained. After feature extraction of three authentications the fusion technique as feature level fusion is used for minimising features. This fusion technique is mainly used for reducing time compression when recognition of the images. Then, using the optimised Artificial Neural Network (ANN) with Particle Swarm Optimisation (PSO) algorithm, the images are classified as recognition and non-recognition. Finally while providing the test images the recognised images are identified for security purpose.