WCSE 2016
ISBN: 978-981-11-0008-6 DOI: 10.18178/wcse.2016.06.113

An Improved Recognition Method of Traffic Gestural Command Based on Skeleton Data

Fan Huang, Qian Huang, Chongwen Wang

Abstract— This paper proposes an improved method based on skeleton data to increase the recognition efficiency and recognition rate of traffic gestural command. First, Kinect is used to obtain the coordinate data of skeleton nodes. Then the cosine value of deflection angle is extracted as feature for training and testing. Finally, DTW (Dynamic Time Warping) algorithm is improved by deflection weighting and sample filtering to recognize the given samples. Experiment shows that the average recognition rate of the method in this paper is up to 97%.

Index Terms— traffic gestural, skeleton data, gesture recognition, DTW.

Fan Huang, Qian Huang and Chongwen Wang
School of Software Engineering, Beijing Institute of Technology, CHINA

[Download]


Cite: Fan Huang, Qian Huang, Chongwen Wang, "An Improved Recognition Method of Traffic Gestural Command Based on Skeleton Data," Proceedings of 2016 6th International Workshop on Computer Science and Engineering, pp. 644-649, Tokyo, 17-19 June, 2016.