• P.vikram
• Gayathridevi
• SRINATH K
• Vishwajit N

Ear recognition has emerged as a robust biometric technique, prized for its immunity to facial expressions and minimal physical interaction requirements. Study unveils a pioneering algorithm harnessing deep convolutional neural networks (CNNs) for ear recognition, shedding light on the network's acquired features. The ear's consistent facial location, notably in profile views, bolsters its suitability as a biometric identifier, with proven efficacy even in distinguishing identical twins. Furthermore, ear segmentation from facial images presents fewer challenges than facial recognition, benefiting from a predictable background. The integration of support vector machine (SVM) classifiers with CNNs has significantly bolstered the robustness and accuracy of ear recognition models by adeptly managing high-dimensional feature spaces. Renowned for their capacity to find optimal hyperplanes for class separation and maximize margin, SVM classifiers enhance generalization and noise resilience, especially on unseen data. With the versatility to handle both linear and non-linear classification tasks through diverse kernel functions, SVMs effectively capture intricate decision boundaries. The successful applications span text classification, image classification, and pattern recognition domains, attributed to their adaptability to diverse data distributions and high-dimensional data handling capabilities.

CNN,SVM

"Human authentication using ear biometrics", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.f575-f579, March-2024, Available :http://www.jetir.org/papers/JETIR2403570.pdf

"Human authentication using ear biometrics", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppf575-f579, March-2024, Available at : http://www.jetir.org/papers/JETIR2403570.pdf