Title:	REAL-TIME HAND DETECTION AND TRACKING TECHNIQUES FOR HUMAN-ROBOT INTERACTION
Author(s):	Chin-Shyurng Fahn, Keng-Yu Chu
ISBN:	978-972-8939-18-2
Editors:	Katherine Blashki
Year:	2010
Edition:	Single
Keywords:	Human-robot interaction, hand detection, hand tracking, gesture recognition
Type:	Full Paper
First Page:	179
Last Page:	186
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	In this paper, we present part of a real-time face tracking together with hand tracking system installed on a human interacting robot. The user can utilize dynamic or static gestures that we currently define to conduct and interact with the robot. For the construction of such a robot vision system, we adopt a PTZ camera which is commonly employed in video surveillance to capture surrounding information. This study acts as the pre-processing operation used for achieving gesture recognition later. In the hand detection and tracking procedure, we propose a new method of mapping a circular plate to locate hands, which allows humans to wear short-sleeved clothes. After a hand region is detected and the centroid point of the hand region is determined, the tracking action will be continuously carried out to acquire a hand motion trajectory by the aid of a local search area around the hand region. At the feature extraction stage, we can find the orientation between two consecutive points from the hand motion trajectory, and it serves as the principal feature for gesture recognition. Experimental results reveal that the correct rate of hand tracking is not less than 98% in general situations and exceeds 94% when the hand is moving quickly. Besides this, we set the image resolution to be 320×240 pixels in our experiments, and the total system execution time is ranged from 60 to120 milliseconds for processing an image frame. The whole system performance is very satisfactory, and we are encouraged to commercialize the robot in the near future.
	Go Back