Three Modals Biometric Information Fusion Based on Image Quality Estimation

This paper addresses a multimodal biometric verification system model based on fingerprint, paim-print and hand-geometry, whose features can be extracted from the human hand and the information fusion is applied at matching score level based on multimodal image quality estimation. The verification process is organized as follows: image capture; processing; sub-images extraction; features extraction; matching score normalization with a new method (a new segment linear function); fusion at matching score level based on image quality, finally a decision made. The system was tested on a database of 100 persons and the experimental results indicate the possibility of the combination.