A novel method of mine target tracking and location based on wireless sensor and visual recognition

Abstract

In view of limited monitor, target occlusion and multipath interference in the mine NLOS environment result in the problem of the target localization precision decrease and delay.This paper proposes a novel fusion method based on wireless sensor and visual recognition for mine moving target tracking and positioning in coal tunnel scene.Firstly, through the radio frequency identification technology (RFID) sensing, the initial location information of the target was accessed, and then the visual tracking of moving targets was applied to capture its current location information.Finally, in order to estimate accurate target position, the target of the sensing information and visual tracking position information were weighted fusion.Experimental results show that the proposed method can improve the positioning accuracy and reduce the location error with less consuming of time.Compared with radio frequency identification technology and visual positioning method, the average positioning error and computation time is reduced by 62.8% and 79.5% respectively.The proposed method has strong robustness in noise environment, effectively realize real-time tracking and precise positioning of the target.

Keywords

• mine
• wireless awareness
• visual recognition
• multi-source information fusion
• target tracking location