Human Face Detection and Segmentation using Eigenvalues of Covariance Matrix, Hough Transform and Raster Scan Algorithms

In this paper we propose a novel method for human face segmentation using the elliptical structure of the human head. It makes use of the information present in the edge map of the image. In this approach we use the fact that the eigenvalues of covariance matrix represent the elliptical structure. The large and small eigenvalues of covariance matrix are associated with major and minor axial lengths of an ellipse. The other elliptical parameters are used to identify the centre and orientation of the face. Since an Elliptical Hough Transform requires 5D Hough Space, the Circular Hough Transform (CHT) is used to evaluate the elliptical parameters. Sparse matrix technique is used to perform CHT, as it squeeze zero elements, and have only a small number of non-zero elements, thereby having an advantage of less storage space and computational time. Neighborhood suppression scheme is used to identify the valid Hough peaks. The accurate position of the circumference pixels for occluded and distorted ellipses is identified using Bresenham-s Raster Scan Algorithm which uses the geometrical symmetry properties. This method does not require the evaluation of tangents for curvature contours, which are very sensitive to noise. The method has been evaluated on several images with different face orientations.

Authors:

• J. Prakash
• K. Rajesh

Keywords:

• Circular Hough Transform
• Covariance matrix
• Eigenvalues
• Elliptical Hough Transform
• Face segmentation
• Raster Scan Algorithm.

References:

[1] R. Chellappa, C.L. Wilson, S. Sirohey, "Human and machine recognition
of faces: A survey", Proc. of IEEE, V83, 1995, pp. 705-740.
[2] A. Samal, P. A. Iyengar "Automatic recognition and analysis of human
faces and facial expressions: A survey", Pattern Recognition, 25(1),
1992, pp. 65-77.
[3] T. Kanade, "Picture processing system by computer complex and
recognition of human faces", PhD thesis, Dept. of Information science,
Kyota university, Japan, Nov-2003.
[4] R. Brunelli, T. Poggio, "Face recognition through geometrical feature".
Proc. European conference on computer vision, 1992, pp 792-800.
[5] B.S. Manjunath, R. Chellappa, C. Vonder Malsburg, "A feature based
approach to face recognition", Proc. of IEEE conference, computer
vision pattern recognition, 1992, pp. 373-378.
[6] A.L. Yuille, D.S. Cohen, P.W. Hallinan, "Feature extraction from faces
using deformable templates", Proc. of IEEE conference, computer vision
pattern recognition, 1989, pp. 104-109.
[7] G V. Govindaraju, S.N. Srihari, D.B. Sher, "A computational model for
face location", Proc. international conf. computer vision, 1990, pp. 718-
721.
[8] M.Turk, A.Pentland, "Eigenfaces for recognition", Journal of Cognitive
Neuroscience, vol.3, No.1, 1991.
[9] M. Turk, A. Pentland, "Face recognition using eigenfaces", Proc. IEEE
Conf. on Computer Vision and Pattern Recognition, 1991, pp. 586-591.
[10] S. A. Sirohey, "Human face segmentation and identification", Technical
Report CAR-TR-695, Center for Automation Research, University of
Maryland, College Park, MD, 1993.
[11] Ismail Haritaoglu, David Harwood, Larry S. Davi, "Multiple people
detection and tracking", ICIAP, 1999, pp. 280-285.
[12] Sangho Park, "Head segmentation and head orientation for pose
estimation of multiple people", SSIAI, 2000, pp 192-196.
[13] D. H. Ballard, "Generalizing the Hough transform to detect arbitrary
shapes", Pattern Recogn., 13(2), 1981, pp. 111-122.
[14] E. R. Davis, "Finding ellipses using the generalized Hough transform",
Pattern Recogn., Feb 1989, pp. 87-96.
[15] Wen-Yen Wu, Mao-Jiun J. Wang, "Elliptical object detection by using
geometric properties", Pattern Recogn, Vol. 26, No. 20, April. 1993, pp.
1499-1509.
[16] Chun-Ta Ho, Ling-Hwei Chen, "A Fast ellipse/circle detector using
geometric symmetry", Pattern Recogn., Vol. 28, No. 1June 1994, , pp.
117-124.
[17] Du-Ming Tsai, H. T. Hou, H.J. Su, "Boundary based corner detection
using Eigen values of covariance matrices", Pattern. Recogn. 20, Sept
1998, pp. 31-40.
[18] D. S. Guru B, H. Shekar, P. Nagabhushan, "A simple and robust line
detection algorithm based on small Eigen value analysis", Pattern
Recogn., 25, Aug 2003, pp. 1-13.
[19] J. Prakash, K. Rajesh, "Extracting geometric primitives: Combined
approach of Hough transform, Eigen values and raster scan algorithms",
International Journal of systemics, Cybernetics and Informatics
(IJSCIJan 2007, ), pp. 48-55.
[20] J. Prakash, K. Rajesh, "elliptical features extraction using eigenvalues
of covariance matrix, Hough transform and raster scan algorithms",
International Journal of computer science and Engineering, vol 1,No 2.
2007, pp. 97-102.
[21] Donald Hearn, M. Pauline Baker, "Computer graphics" (2nd ed),
Pearson Education, 2003.
[22] Rafael C. Gonzalez, Richard E. Woods, "Digital Image processing" (5th
ed), Addison Wesley, 2000.