About some years ago, several biometric technologies are considered mature enough to be a new tool for security and ear-based person identification is one of these technologies. This technology provides a reliable, low cost and user-friendly viable solution for a range of access control applications. In this paper, we propose an efficient online personal identification system based on ear images. In this purpose, the identification algorithm aims to extract, for each ear, a specific set of features. Based on Gabor filter response, three ear features have been used in order to extract different and complementary information: phase, module and a combination of the real and imaginary parts. Using these features, several combinations are tested in the fusion phase in order to achieve an optimal multi-representation system which leads to a better identification accuracy. The obtained experimental results show that the system yields the best performance for identifying a person and it is able to provide the highest degree of biometrics-based system security.