Feature level fusion approach for personal authentication in multimodal biometrics

Biometric is an automated process of identifying or verifying an individual based upon his or her behavioral or physical characteristics. Biometrics speaks loud in the Area of Security, Banking and Forensic department. Single modality based recognition verification is not very robust while combining information from various biometric modalities provides better performance. In our Proposed work Modalities such as Finger print, Palm print and Finger Knuckle prints are used for authenticate a personal. Grey Level Co Occurrence Matrix (GLCM) feature extraction Technique is used to extract the unique characteristics of these Modalities. Extracted Features are then fused in Feature level. During Classification, use of optimized Artificial Neural Network (ANN) with particle swarm optimization (PSO) algorithm recognizes a person with high level security, specificity and sensitivity.