Convolutional Neural Network Based Implementation of Human Ear Recognition System

Human ear recognition is a biometric identification technique which uses unique features of an ear to identify an individual. With the help of ear recognition system, we can add an extra layer of security to the existing system of facial recognition or fingerprint recognition. The structure of the ear remains fairly the same and doesn't change much with aging as compared to face and it also doesn't need to be in close proximity as required in case of fingerprint recognition which makes it ideal tool for passive identification and verification. In this paper, for ear detection, we have used Convolutional Neural Network (CNN) algorithm which is particularly used for image recognition. It consists of three layers namely convolutional layer, pooling layer and fully-connected layer. We aim to improve the robustness and accuracy of the biometric identification technique using CNN. The technique has been applied on own created dataset.