Ear Recognition by using Neural Networks

Abstract: Using ears in identifying people has been interesting at least 100 years. The researches still discuss if the ears are unique or unique enough to be used as biometrics. Ear shape applications are not commonly used, yet, but the area is interesting especially in crime investigation. In this paper, the basics of using ear as biometric for person identification and authentication are presented. In addition, the error rate and application scenarios of ear biometrics are introduced. A set of 17 people has been used for experiments having six or more images each. The data used are given by National Institute of Standards and Technology (NIST). The correct recognition rate is ranging between 84.3 % and 91.2 % for artificial neural network matching. It depends on neural network training parameters