Biometric Recognition Using Three-Dimensional Ear Shape
Kevin W. Bowyer
Department of Computer Science and Engineering
University of Notre Dame, IN 46556
Previous works have shown that the ear is a promising candidate for biometric identification. However,
in prior work, the pre-processing of ear images has had manual steps, and algorithms have not neces-
sarily handled problems caused by hair and earrings. We present a complete system for ear biometrics,
including automated segmentation of the ear in a profile view image and 3D shape matching for recog-
nition. We evaluated this system with the largest experimental study to date in ear biometrics, achieving
a rank-one recognition rate of 97.8% for an identification scenario, and equal error rate of 1.2% for a
verification scenario on a database of 415 subjects and 1,386 total probes.
Keyword: biometrics, ear biometrics, 3-D shape, skin detection, curvature estimation, active contour,
iterative closest point.
Ear images can be acquired in a similar manner to face images, and a number of researchers have
suggested that the human ear is unique enough to each individual to allow practical use as a biometric.
Several researchers have looked at using features from the ear’s appearance in 2D intensity images [6,
16, 5, 27, 17, 10, 11, 23, 31]. A smaller number of researchers have looked at using 3D ear shape [8, 4].
Our own previous work that compared ear biometrics using 2D appearance and 3D shape concluded that
3D shape matching allowed greater performance . In other previous work, we compared recognition
using 2D intensity images of the ear with recognition using 2D intensity images of the face and suggested
that they are comparable in recognition power [6, 27]. Also, ear biometric results can be combined with
results from face biometrics. Thus, additional work on ear biometrics has the promise to lead to increased
recognition flexibility and power in biometrics.
This paper builds on our previous