Hand Detection Based on Rectangle Features and Improved Adaboost

【Abstract】In order to detect and recognize hand gesture from video sequence in real time accurately, a set of rectangle features are used to describe the hand characteristics, and the fast calculation methods of their features and the evaluation ways of the separability of hand gesture class are given. The Adaboost algorithm is improved to deal with the excessive training. Experimental results show that rectangle features can obtain the reliable detector and has a good real-time performance and adaptive capacity in complex backgrounds, but it is more sensitive to the changes of hand gesture. When hand gesture rotates small angle, the detection rates are 95% above. 【Key words】rectangle features; improved Adaboost; complex ground; hand gesture detection; hand gesture recognition

[1]  Huang Chang,et al.  A Multi-View Face Detection Based on Real Adaboost Algorithm , 2005 .

[2]  Mihir Bellare,et al.  Entity Authentication and Key Distribution , 1993, CRYPTO.

[3]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[4]  Mihir Bellare,et al.  Random oracles are practical: a paradigm for designing efficient protocols , 1993, CCS '93.

[5]  Mathias Kölsch,et al.  Robust hand detection , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..

[6]  Alfred Menezes,et al.  Key Agreement Protocols and Their Security Analysis , 1997, IMACC.

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

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