DocumentCode :
2316603
Title :
An Object Tracking System Based on Human Neural Pathways of Binocular Motor System
Author :
Zhang, Xiaolin
Author_Institution :
P & I Lab, Tokyo Inst. of Technol.
fYear :
2006
fDate :
5-8 Dec. 2006
Firstpage :
1
Lastpage :
8
Abstract :
The human eyes cannot easily focus on two separate targets at the same time. The two eyes have their own cooperative movements called conjugate and vergence eye movements. The reason for the cooperative movements and a method to achieve binocular movements using active cameras are discussed in this paper. In order to achieve realistic eye movements, we proposed a binocular motor system model based on the human neural pathways of the binocular motor system. Using this model, an active camera control system was constructed. The system exhibited several characteristics specific to human-eye movement, including the following: (1) both cameras moved in tandem and had the same target point in the central area. This characteristic was considered a basic condition for structuring a stereo-image using the image signals from both eyes; (2) if one camera was obstructed by an obstacle, it would follow the movement of the other camera to find the target promptly when the obstacle was removed; (3) this system had the ability to compensate for the blurring caused by base movement
Keywords :
image motion analysis; object detection; active camera control system; binocular motor system model; binocular movement; cooperative movement; human neural pathway; image signal; object tracking system; stereo-image; vergence eye movement; Cameras; Control system synthesis; Eyes; Head; Humans; Neural pathways; Optical computing; Retina; Robot kinematics; Robot vision systems; Binocular motor system; Conjugate and Vergence eye movement; Robotics; Smooth Pursuit; Vestibulo-Ocular Reflex;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control, Automation, Robotics and Vision, 2006. ICARCV '06. 9th International Conference on
Conference_Location :
Singapore
Print_ISBN :
1-4244-0341-3
Electronic_ISBN :
1-4214-042-1
Type :
conf
DOI :
10.1109/ICARCV.2006.345126
Filename :
4150036
Link To Document :
بازگشت