A Novel Multimodal Biometrics System with Fingerprint and Gait Recognition Traits Using Contourlet Derivative Weighted Rank Fusion

We propose a novel multimodal biometric system that uses fingerprint and gait recognition traits. Fingerprint images are gathered with the help of a touchless optical sensor with total internal reflection and a capacitive line sensor. Gait samples are obtained as continuous images in 3 frames from a camera in burst mode. Features from these images are extracted with the help of Minutia Cylinder-Code (MCC) minutia descriptor. Further, the images are compared with the gathered sample database of 80 sample images. The comparison score is normalized using geometric mean. Fusion of the normalized images is done using contourlet derivative weighted rank fusion. A combination of these techniques provides an improved performance for authentication using biometrics.