Ear Recognition by means of a Rotation Invariant Descriptor

Iannarelli´s studies showed that ear shape can be considered a biometric identifier able to authenticate people as well as more established biometrics like face or voice, for instance. However, very few researches can be found in literature about ear recognition. In most cases techniques already working in other biometric fields, such as PCA (principal component analysis), are applied to ear. Eigen-ears provide high recognition rate only in closely controlled conditions. Indeed, even a slight amount of rotation can cause a significant drop in system performance and in unattended systems rotations occur very frequently. In this paper, we propose the use of a rotation invariant descriptor, namely GFD (generic Fourier descriptor), to extract meaningful features from ear images. This descriptor results to be quite robust to both ear rotations and illumination changes. Experimental results confirm the superiority of this approach even compared to Eigen-ears