Ear Biometric: A Deep Learning Approach

CNN approach using deep learning had been applied in the current study for the detection of ear biometrics. Real-time ear pictures from Google were trained in the deep CNN model to conduct this experiment. The dataset was initially improved, preprocessed, and divided for training and validation by using MATLAB. For the training and validation, the dataset is split in the ratio of 80:20. The obtained datasets were also used to train and validate the deep CNN model. Following the implementation of the deep CNN model, the total accuracy acquired was 99.45%. This experiment was conducted to raise research awareness on biometrics and to create a useful model for ear biometrics.