Multimodal recognition using face profile and ear

Ear recognition is proved to be a new and promising subject in biometrics authentication. As the ear is located at the side face, it is reasonable to combine ear with face profile for recognition in such scenarios as frontal face images are not available. Face profile is an important aspect of face recognition, it provides a complementary structure of the face that is seen in non-profile view. In a profile image, the profile silhouette and the ear are two biometrics that can be used for authentication. In this paper, the face profile silhouette is combined with the ear to form a multimodal image, and a Full-space Linear Discriminant Analysis (FSLDA) is applied for recognition. Experiments are performed on USTB multimodal image database. The recognition rate of 96.2% on 79 subjects proves the efficiency of FSLDA method and the feasibility of ear and face profile combined image for personal authentication.