Title:
|
FUZZY ACTIVE CONTOURS APPLIED TO EAR BIOMETRICS |
Author(s):
|
José F. Vélez, Ángel Sánchez, A. Belén Moreno, Shamik Sural |
ISBN:
|
978-989-8533-14-2 |
Editors:
|
Hans Weghorn and Pedro Isaías |
Year:
|
2012 |
Edition:
|
Single |
Keywords:
|
Biometrics; verification; ear images; active contours; fuzzy model; control access system. |
Type:
|
Full Paper |
First Page:
|
235 |
Last Page:
|
242 |
Language:
|
English |
Cover:
|
|
Full Contents:
|
click to dowload
|
Paper Abstract:
|
Biometric recognition of people using their ear images as patterns has recently received a significant attention. This can be due to the fact that ear images present some high discriminative features and also they do not have some problems associated with other biometric modalities. Ear images are less intrusive than other biometric modalities, and they have no variability due to expressions or pose like it happens with faces. This paper describes a prototype of verification system that uses outer ear images as patterns. We apply fuzzy snakes-based approach for efficiently solving some of the practical constraints in the ear verification problem. Our method has the advantage of using only one learning pattern per person (i.e. the snake that is created for each individual through the automatic delineation of his/her ear image). Next, the verification process is organized in two stages. First, the snake of a person is adjusted to the test ear image through an iterative process. In the second stage, some measures extracted from the quality of the snake adjustment are used to decide the acceptance or not of the test ear image. The low FAR error (around 7%) achieved by our system make it suitable for a control access application. |
|
|
|
|