DocumentCode :
130086
Title :
A magnetic-assistant multi-ocular tracking system
Author :
Qing He ; Ning Wei ; Chao Hu ; Meng, Max Q.-H. ; Youpan Hu
Author_Institution :
Guangzhou Inst. of Adv. Technol., Guangzhou, China
fYear :
2014
fDate :
28-30 July 2014
Firstpage :
716
Lastpage :
720
Abstract :
This paper proposes a hybrid tracking approach which combines the optical and magnetic localization technologies. It first describes the collinear epipolar plane model, then rewrite and enhance the epipolar constraint under this model. Based on it, this paper further discusses the point corresponding and target recognition in multi-ocular infrared tracking systems. Aiming at the calculation complexity during these two processes, it employs magnetic localization technology to tag and roughly localize the tracked targets and then get refined results by optical localization, which dramatically reduces the calculated amount in tracking.
Keywords :
infrared detectors; magnetic sensors; optical tracking; target tracking; collinear epipolar plane model; epipolar constraint; hybrid tracking system; magnetic assistant multiocular tracking system; magnetic localization technology; multiocular infrared tracking system; optical localization technology; point corresponding process; target recognition; Cameras; Magnetic flux; Magnetic resonance imaging; Magnetic sensors; Optical imaging; Target recognition; Target tracking; angle corresponding criterion; magnetic localization; perfect corresponding set; target recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information and Automation (ICIA), 2014 IEEE International Conference on
Conference_Location :
Hailar
Type :
conf
DOI :
10.1109/ICInfA.2014.6932745
Filename :
6932745
Link To Document :
بازگشت