Iris Recognition Using Combined Feature Vector

Abstract

Biometric recognition system is a reliable authentication to identify a living person based on physiological or behavioral characteristics system. Among the existing biometrics Iris biometric is widely accepted and is having high recognition accuracy due to its unique special features. The iris incorporates fine texture and even genetically similar people have entirely independent iris textures. In this paper, Iris biometric system is considered to develop a recognition model. Various texture features are extracted using Local Binary Pattern (LBP), Local Gabor XOR Pattern, Gray Level Co-occurrence matrix, Gabor features to form a combined feature vector in order to train the Neural Network using Bat algorithm. The performance of recognition model is analyzed by using sample database.

Keywords: LBP, GLCM, Gabor , LGXP, Bat