Inertia-Based Ear Biometrics: A Novel Approach

The human ear has been deemed to be a source of data for person identification in recent years. Ear biometrics has distinct advantages, such as visibility from a distance and ease with which images could be captured. This paper elaborates on a novel approach to ear biometrics. We propose moment of inertia-based biometric for the ears in any random orientation. The features concerned are the moment of inertia about the major and minor axes, corresponding radii of gyration, and the planar surface area of the ear. The databases of the said features were collected through ear images of 600 subjects. Principal component analysis of the features demonstrated that the radius of gyration with respect to the major axis, moment of inertia about the minor axis, and radius of gyration about the minor axis are significant attributes contributing to major variability. The person identification system developed showed recognition rates of 99% with just three attributes, when compared with the 96% recognition rate when all five attributes were considered. The evaluation of the system was done on several metrics. All metrics were found to be insignificant in their magnitude, which is suggestive of robustness and excellent authentication performance.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.