Ear Biometrics Using Deep Learning: A Survey

This paper explores ear biometrics using a mixture of feature extraction techniques and classifies this vector deep learning with convolutional neural network. exploration uses images from 2D facial profiles images. The investigated are Zernike Moments, local binary pattern, Gabor filter, Haralick texture moments. normalised is used to examine whether network better at identifying the than other commonly machine techniques. widely that were compare them decision tree, naïve Bayes, K-nearest neighbors (KNN), support (SVM). proved bag classification technique was standard result achieved by 92.00% average identification rate for both left right ears.