Comparison of traditional haar classifiers used in face detection applications with an alternative classifier for four stages filtering

In this study, Haar-like features used for face detection on images and adaboost algorithm have been introduced, and an additional distinctive feature of traditional Haar features have been proposed. The performances of recommended feature and traditional Haar features in `CBCL-MIT Center For Biological and Computation Learning' face database on the first 200 containing faces positive images and on the first 400 containing nonfaces negative pictures have been tested and the success rates (%) have been indicated in a table. As a result, the success rate of the supposed feature on face photos in the first three stages have been much higher than traditional Haar features.

[1]  Luhong Liang,et al.  A detector tree of boosted classifiers for real-time object detection and tracking , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[2]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[3]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[4]  Paul A. Viola,et al.  Robust Real-time Object Detection , 2001 .

[5]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[6]  Takeshi Mita,et al.  Joint Haar-like features for face detection , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[7]  Rainer Lienhart,et al.  An extended set of Haar-like features for rapid object detection , 2002, Proceedings. International Conference on Image Processing.