Ear Detection Based on Skin-Color and Contour Information

Ear detection is a challenging task in the ear recognition system. In this paper a tracking method which combines both skin-color model and intensity contour information is proposed to detect and track the human ear in a sequence frames. As a fast computationally efficient algorithm on color-based tracking, a modified CAMSHIFT algorithm is applied to roughly track the profile image as the region of interest (ROI) and then contour fitting is operated on ROI for further accurate location due to the ample contour information the ear contains. Experiment results show that this method can meet the real-time requirement and works well in practical situations.