On an Ear Recognition Method Based on Principal Component Analysis and Sphere Support Vector Machine
Rong Li · Journal of Southwest China Normal University · 2013
An ear recognition method has been proposed in this paper on the basis of principal component analysis and hyper-support vector machine.First an ear has been detected from an image of side face,and then the PCA method used to extract ear features,and,at last,the ear features have been trained and tested by means of sphere SVM.Comparing with other traditional multi-class SVMs,the proposed method achieved higher performance,thus,there comes a novel effective approach to the non-intrusive biometric recognition.