Soft and hard biometrics fusion for improved identity verification

Traditional biometric systems that are based on single biometric usually suffer from problems like unacceptable error rates, spoof attacks and insufficient population coverage. To improve system's reliability, two or more biometrics of the same identity could be combined in a manner that enhances the efficiency of the system. Multi-biometric systems integrate typical physiological (fingerprint, iris, face...etc.) and behavioral (e.g. gait, voice, signature...etc.) traits. However, other features like color of skin, color of irises and height can also be included to enhance the overall performance. In this paper, a framework for integrating a soft biometric like the color of a human iris within a multi modal biometric system that combines hard biometrics such as fingerprint and iris texture. Steerable pyramid filters and multi-channel log-Gabor filters are suggested here for extracting proper features of the fingerprints and iris textures respectively. Experimental results, considering two different fusion strategies, have shown a significant improvement in the performance compared to the results of the standalone combined fingerprint and iris matchers.