Copyright © 2012 Xiaoyan Wang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This paper is concerned with the recognition of dynamic hand gestures. A method based on Hidden Markov Models (HMMs) is presented for dynamic gesture trajectory modeling and recognition. Adaboost algorithm is used to detect the user's hand and a contour-based hand tracker is formed combining condensation and partitioned sampling. Cubic B-spline is adopted to approximately fit the trajectory points into a curve. Invariant curve moments as global features and orientation as local features are computed to represent the trajectory of hand gesture. The proposed method can achieve automatic hand gesture online recognition and can successfully reject atypical gestures. The experimental results show that the proposed algorithm can reach better recognition results than the traditional hand recognition method.