Localization of Ear Using Outer Helix Curve of the Ear

This paper presents an efficient approach for localization of ear from an arbitrary 2D side face image with varying background. Outer helix curves of ears moving parallel to each other are used as feature for localizing ear in an image. Using Canny edge detector, edges are extracted from the whole image. These edges are segmented in convex and concave edges. From these segmented edges expected outer helix edges are determined after eliminating non-ear edges. Final outer helix edge of an ear is constructed using expected outer helix curves. Decision is made on a constructed curve whether it belongs to outer helix of ear or not. This technique is implemented on IITK, India database containing 700 samples. Accuracy of localization is more than 93%

[1]  L. Mazorra,et al.  Fitting ear contour using an ovoid model , 2005, Proceedings 39th Annual 2005 International Carnahan Conference on Security Technology.

[2]  Wilhelm Burger,et al.  Ear biometrics in computer vision , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[3]  Roland T. Chin,et al.  Deformable contours: modeling and extraction , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Mark S. Nixon,et al.  Automatic ear recognition by force field transformations , 2000 .

[5]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Mark S. Nixon,et al.  A new force field transform for ear and face recognition , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[7]  M. Burge,et al.  Ear Biometrics , 1998 .

[8]  Hui Chen,et al.  Human ear detection from side face range images , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..