Convolution Neural Network Based Method for Biometric Recognition

In recent years, there has been increasing interest in the potential of precisely identifying individuals through ear images within the biometric community, owing to the distinctive characteristics of the human ear. This paper introduces deep neural network architecture for ear recognition. The suggested method incorporates a preprocessing stage that enhances significant features in ear images through contrast-limited adaptive histogram equalization. Subsequently, a classifier with deep convolutional neural network is employed to recognize the preprocessed ear images. Experimental results demonstrate a remarkable testing accuracy of 97.92% for the proposed recognition system.