Implementing Feature Extraction and GA based Feature Selection for Fingerprint-Iris and Fingerprint-Face

To identify the physiological and behavioral parameters of a specific person, biometrics are used. It implements several characteristics such as faces, iris, fingerprints or retinas to capture the identity. A multimodal biometric system uses a combination of two or more traits that are not forgotten, copied or stolen. Several discriminant features are extracted from samples which is represented in a feature vector. The resultant feature vector is of the order of high dimensionality which causes computational complexity and impairs the performance of classifiers. To rectify this, feature selection methods are introduced to obtain a subset of optimal features. This paper presents a multimodal biometric framework based on fingerprint, iris and face. Feature extraction is done by Gabor filter, Local Tetra Pattern and feature selection is done by Genetic Algorithm (GA).