A BI-MODAL PERSON IDENTITY AUTHENTICATION USING FACE AND GAIT BIOMETRICS

ABSTRACT

Biometric authentication of a person is highly challenging and complex problem. Face and gait Biometrics have received significant attention in computer vision field over the years. However, these modes of identification have challenges due to their large varying appearances and high complex pattern distributions. It is noted that the complementary properties of these two biometrics suggest fusion of them. Face identification is more reliable when the person is close to the camera. On the other hand, gait is a suitable Biometric trait for human identification at a distance. In this research, we hope to carry out the procedures of human recognition using face and gait biometrics in order to increase the security level of the application areas. With the fusion of the two biometrics, we can minimize the system false acceptance rates. This work made use of the fusion of face and gait biometrics for person authentication. Features extracted from face and gait images are used to train eigenface and eigengait models. The fusion of these models at the decision level is carried out. The experimental results expose the effectiveness of the proposed method against unimodal identification methods due to the better accuracies achieved.

Keywords: eigenface, eigengait, PCA, gait, face biometrics