Ear Biometrics-Based Security Feature Extraction: A Deep Learning Model

It has long been argued that the human ear has an identification capacity. It has been demonstrated that each person’s ear is distinctive and can be utilized as a biometric to get beyond the drawbacks of the current biometrics. An approach to using the ear as a biometric for identification is presented in this paper. This study applies a deep learning-based convolutional neural network model to a database that has been digitally transformed, and the findings are encouraging. In this paper, we have used various pre-processing methods using different variations of Pa compared to the above methods. These methods have been done harmoniously with selected feature extraction methods, bringing outstanding results and promise. We have used deep neural network models for different ear characteristics to train a robust model, which is a stepping stone for future research areas and helps us develop a better understanding and application for ear biometrics.