An Efficient Ear Biometric Authentication System

The Previous research has demonstrated human ears can be used for identification, gender recognition, age identity and family relationship identification [1] and examined whether a person's ears are harmonious. The symmetry performances, though, haven't exactly been stellar. Our paper amplifies the identification, investigation of gender classification on ear images and authenticates an individual based on ear images in both cases pointing to decide the ear parts from which acknowledgment is determined. The model-based technique was discussed in detail and In our paper we talked about the HOG and the morphological algorithm, which we had applied. Finally we investigated whether convolutional neural network that is deep learning outperforms other well-known machine learning methods at identifying the ear This paper revealed that classification technique of deep learning using CNN was superior than Machine Learning techniques. The method to extract features from the dataset images and match with the trained images of the database has been finished using Python.