• Tejal Nanaware
• Siddharth Nandargi
• Kavya Parag
• NehaPote
• ChandanSingh Rawat

Biometric technology has always been about innovation and new technology. It was the world of passwords and identity proofs that ruled the world of security before biometrics happened. Ear biometric is invariant from childhood to early old age as compared to other biometrics. Ear biometric is a non-invasive type of recognition. This paper presents a delineated comparison between two techniques which are used for recognition based on various characteristics of the ears. Use of SURF features along with K NearestNeighbours classification technique has been incorporated along with the training of Convolution Neural Network model where the features extracted in the first few layers are classified in the last layer. The experimental results show the relationship between computational time taken and varying number of features for different scenarios with the aid of KNN classifier, also the proposed approach of recognition by training CNN model gives an accuracy of 86% on AWE database.

Ear biometrics, convolutional neural network, SURF, K-NN classifiers

"EAR BIOMETRIC TECHNIQUES: A COMPARATIVE APPROACH ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.102-106, May 2019, Available :http://www.jetir.org/papers/JETIRCS06025.pdf

"EAR BIOMETRIC TECHNIQUES: A COMPARATIVE APPROACH ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp102-106, May 2019, Available at : http://www.jetir.org/papers/JETIRCS06025.pdf