Ear detection using active contour model

Ear detection in face images, which is invariant to background and pose, is very challenging. This paper proposes a novel solution to this problem with the use of Snakes as active contour model. The proposed method encompasses two stages, namely, Snake-based Background Removal (SBR) and Snake-based Ear Localization (SEL). SBR is used to remove background from a face image and thereafter, SEL is used to localize the ear. A feedback is provided to this system with the use of Support Vector Machines (SVM) which classifies the region of the image localized by SEL as an ear or non-ear entity. Experimental results obtained by applying the proposed algorithm on two benchmark face databases, namely CMU-PIE and ColorFERET, show a significant improvement in Localization Rate, and a substantial reduction in the computation time.