Multimodal Biometric Recognition

Biometric systems make use of the physiological or behavioral traits of individuals, for recognition purposes. These traits include fingerprints, hand-geometry, face, voice, iris, retina, gait, signature, palm-print, ear, etc. Biometric systems that use a single trait for recognition (i.e., unimodal biometric systems) are often affected by several practical problems like noisy sensor data, non-universality and/or lack of distinctiveness of the biometric trait, unacceptable error rates, and spoof attacks. Multimodal biometric systems overcome some of these problems by consolidating the evidence obtained from different sources. Researchers have shown that the use of multimodal biometrics provides better authentication performance over unimodal biometrics. Biometric fusion can be performed at image level, feature level, match score level, decision level, and rank level.

We have developed a multimodal biometric system that efficiently combines fingerprint, iris and palmprint recognition. Extracted features are combined and a final score is computed for classification. Code has been tested with CASIA Iris Image Database Version 1.0 and CASIA Palmprint Image Database. Fingerprint database used in our experiments was a collection of fingerprint images taken with an UPEK swipe fingerprint reader with capacitive sensor and USB 2.0 connection. Database is 16 fingers wide and 8 impressions per finger deep (totally 128 fingerprints). Other biometric modalities are available on request.