Hybridization of Bimodal Biometrics for Access Control Authentication

Abstract

Single biometric trait for authentication is widely used in some application areas where security is of high importance. However, biometric systems are susceptible to noise, intra-class variation, non-universality and spoof attacks. Thus, there is need to use algorithms that overcome all these limitations found in biometric systems. The use of multimodal biometrics can improve the performance of authentication system. This study proposed using both fingerprint and face for authentication in access system. The study integrated fingerprint and face biometric to improve the performance in access control system. Fingerprint biometric, this paper considered restoration of distorted and misaligned fingerprints caused by environmental noise such as oil, wrinkles, dry skin, dirt, displacement etc. The noisy, distorted and/or misaligned fingerprint produced as a 2-D on x-y image, is enhanced and optimized using a hybrid Modified Gabor Filter-Hierarchal Structure Check (MGF-HSC) system model. In face biometric Fast Principal Component Analysis (FPCA) algorithm was used in which different face conditions (face distortions) such as lighting, blurriness, pose, head orientation and other conditions are addressed. The algorithms used improved the quality of distorted and misaligned fingerprint image. They also improved the recognition accuracy of distorted face during authentication. The results obtained showed that the combination of both fingerprint and face improve the overall performance of biometric authentication system in access control.