FUZZY CLASSIFICATION, IMAGE SEGMENTATION AND SHAPE ANALYSIS FOR HUMAN FACE DETECTION

In this paper, we describe a new algorithm for human face detection. The algorithm uses both color and shape analysis. Color analysis is based on the novel idea of fuzzy classification that manipulates ambiguity in colors; using this technique we can robustly classify skin region and non-skin region. In order to decide whether the skin region is a face or not, we test its shape by applying an edge detection using Canny operator and an ellipse detection using Hough transform. We show with experimental results how the important detection ratio makes our system an effective tool for face detection

[1]  H. D. Cheng,et al.  A fuzzy logic approach to image segmentation , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[2]  Charles A. Bouman,et al.  A simple and efficient face detection algorithm for video database applications , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[3]  Qiang Ji,et al.  A new efficient ellipse detection method , 2002, Object recognition supported by user interaction for service robots.

[4]  Narendra Ahuja,et al.  Detecting Faces in Images: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Edward J. Delp,et al.  An unsupervised color image segmentation algorithm for face detection applications , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).