상세정보
학술대회
Moving Object Performance Analysis System Using Multi-camera Video and Position Sensors
- 발행일
- 202002
- 출처
- International Conference on Big Data and Smart Computing (BigComp) 2020, pp.441-445
- 협약과제
- 19HH2500, 보행자 위치공간 인지 증강 및 스포츠 경기력 분석을 위한 정밀측위 원천기술개발, 박상준
- 초록
- We present an moving object performance analysis system using multi-camera video and position sensors. The proposed system firstly adopted a multi-camera video based tracking technique with moving object identification. Moreover, technology for more precise tracking and identification by using multiple camera and IR-UWB fusion technology. In the IR-UWB based tracking, One-way ranging technology is used to estimate the positions of the players in real-time. In the test experiment, the proposed system showed the performance of average tracking speed of 16 fps with four cameras the average player identification rate as 98.8 % and the average position error is 0.48 meters.
- KSP 제안 키워드
- Fusion technology, Identification rate, Impulse Radio Ultra-Wideband(IR-UWB), Moving Object, Multi-camera, One-way ranging, Performance analysis, Position error, Real-Time, Tracking speed, Tracking technique