ABSTRACT
In contemporary society, there is a specific need to identify a person accurately for safety and security from a social distance because of the present prevailing COVID-19 situation. Hence, instantly recognising a person using non-intrusive ear biometric system have been recently attracted the attention of the biometric community because of its stable features with age and its ability in identifying the twins. As per the knowledge, there is no real time (Commercial) deployment of ear biometric system. Hence, this paper presents the design and implementation of an ear biometric system using the Texas Instruments DM6437 Evaluation Module (EVM). The ear biometric system has been implemented using Principle Component Analysis (PCA) with a suitable distance metric based on the image distribution function. The performance of the developed ear biometric system is analysed using typical database from research laboratory. From the experimental results it is evident that the designed ear biometric system, with City block distance metric, shows an improvement of 4.3% in recognition rate when compared with the existing system.