A Passive and Multimodal Biometric System for Personal Identification

M. Masudur Rahman, R. Hartley (Australia), and S. Ishikawa (Japan)


Eigenface, eigenear, multimodal biometric, passive biometric, PCA algorithm.


A passive and multimodal biometric system for personal identification using human faces and ears is introduced in this paper. We have applied the PCA approach to images of the face and right ears using the same set of objects. A new technique of tracking a human face and an ear from the same image of a particular person is proposed that provides us to use a single camera for developing the proposed biometric system. The proposed system uses the extracted face and ear images to develop the respective feature spaces via PCA algorithm called eigenfaces and eigenears, respectively. A new face and ear can be classified by calculating the Euclidean distance between the classes of eigenfaces or eigenears and the new face or ear, respectively. Eighteen persons’ faces and ears are employed for developing the databases of the eigenfaces and eigenears, and new faces and ears taken from various sessions of the same persons are employed for the identification. The proposed multimodal biometric system shows promising results than individual face or ear biometrics investigated in the experiments.

Important Links:

Go Back