Face Detection Using Ellipsoid Skin Model

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 236)

Abstract

In the presence of unequal lighting conditions and complex backgrounds, this paper proposes a novel face detection algorithm for color images which consists of four pivotal parts primarily: image preprocessing based on color balance and light equalization, skin region segmentation and extraction based on CbrCbgCgr ellipsoid skin model, image post-processing based on morphology, as well as face and facial feature detection based on AdaBoost classifier and facial geometry. Experimental results demonstrate that the algorithm can be effectively applied to the cases of unequal light, complex background and multi-face conditions.

Keywords

Face detection Face recognition Skin model 

References

  1. 1.
    Lam, K.M., Hong, Y.: Location and extracting the eye in human face images. Pattern Recognit. 5, 771–779 (1996)Google Scholar
  2. 2.
    Maio, D., Maltoni, D.: Real-time face location on gray-scale static images. Pattern Recognit. 9, 1525–1539 (2000)CrossRefGoogle Scholar
  3. 3.
    Hsu, R.L., Mottaleb, M.A., Jain, A.K.: Face detection in color images. IEEE Trans. Pattern Anal. Mach. Intell. 5, 696–706 (2002)Google Scholar
  4. 4.
    Liu, L.Y., Sang, N., Yang, S.Y., Huang, R.: Real-time skin color detection under rapidly changing illumination conditions. IEEE Trans. Electron Devices 3, 1295–1302 (2011)Google Scholar
  5. 5.
    Choi, S.I., Jeong, G.M.: Shadow compensation using fourier analysis with application to face recognition. IEEE Signal Process. Lett. 18, 23–26 (2011)CrossRefGoogle Scholar
  6. 6.
    Liu, Q., Peng, G.Z.: A robust skin color based face detection algorithm. In: Proceedings of 2010 2nd International Asia Conference Informatics in Control, Automation and Robotics, pp. 525–528 (2010)Google Scholar
  7. 7.
    Li, Q., Ji, H.B.: Face detection in complex background based on Gaussian models and neural networks. In: Proceedings of International Conference on Signal Process. (ICSP’ 06) (2006)Google Scholar
  8. 8.

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.Department of Computer ScienceChina West Normal UniversityNanchongChina

Personalised recommendations