Real-Time Hand Gesture Recognition Based on Vision

In vision-based hand gesture recognition, an accurate description of gestures and a proper classifier which is chosen for classifying will bring great effect on the result of classification. In this paper, we choose Normalized Moment of Inertia (NMI) and Hu invariant moments of gesture pictures as the features of gestures, and Support Vector Machine (SVM) as the classifier. We experiment the method on eight gestures after training, and get an ideal accuracy of gesture recognition which closes to 97%.

[1]  Ming-Kuei Hu,et al.  Visual pattern recognition by moment invariants , 1962, IRE Trans. Inf. Theory.

[2]  Alois Knoll,et al.  A Naïve Bayes Classifier with Distance Weighting for Hand-Gesture Recognition , 2008 .

[3]  Yen-Ting Chen,et al.  Multiple-angle Hand Gesture Recognition by Fusing SVM Classifiers , 2007, 2007 IEEE International Conference on Automation Science and Engineering.

[4]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[5]  Miroslaw Pawlak,et al.  On the Accuracy of Zernike Moments for Image Analysis , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Chun-Liang Tung,et al.  Dynamic hand gesture recognition using hierarchical dynamic Bayesian networks through low-level image processing , 2008, 2008 International Conference on Machine Learning and Cybernetics.