Eyebrow Segmentation Based on Binary Edge Image
Eyebrow is one of the most salient face features. It has a lot of potential applications in face recognition, nonverbal communication, and so on. In this paper, a novel eyebrow segmentation method based on binary edge image (BEI) is proposed. Our method firstly extracts BEI from a grayscale face image, and then connections between different face components in a BEI are removed using a specially designed algorithm. After that, some eyebrow-analogue segments are extracted from a BEI based on the geometrical property of eyebrows. The fourth step is to locate eyebrows using integral projection approach. Finally the perimeter of an eyebrow block is extracted to finish the segmentation of an eyebrow. Experimental results on a set of 517 AR images with different facial expression and illumination show that a correct eyebrow segmentation rate of 93.4% is achieved, indicating that the proposed method is robust to facial expression and illumination changes.
KeywordsEyebrow segmentation Binary Edge Image (BEI) connection removal illumination change expression change
Unable to display preview. Download preview PDF.
- 3.Dong, Y., Woodard, D.L.: Eyebrow Shape-based Features for Biometric Recognition and Gender Classification: A Feasibility Study. In: 2011 International Joint Conference on Biometrics, pp. 1–8. IEEE Press, New York (2011)Google Scholar
- 4.Kapoor, A., Picard, R.W.: Real-time, Fully Automatic Upper Facial Feature Tracking. In: 5th International Conference on Automatic Face and Gesture Recognition, pp. 8–13. IEEE Press, New York (2002)Google Scholar
- 9.Martinez, A.M., Benavente, R.: The AR Face Database. CVC Technical Report #24 (1998)Google Scholar