Video-based vehicle speed measurement is an important method of traffic accident analysis, but many factors influence its accuracy. The time-interpolation method uses feature length on vehicle body to analyze vehicle speed. The influence of video resolution, vehicle running conditions, angle between the direction of vehicle and camera axis, frame image size in the video, and the vehicle speed were analyzed through 15 vehicle experiments. Error quantization models were obtained. We found that results calculated with high-resolution videos are 0.7% more accurate than results calculated with low-resolution, and that longer feature length displays in videos yielded results with higher accuracy. The error of calculation result is smallest when the camera axis is perpendicular to the vehicle direction. Vehicle speed has little influence on precision. Larger acceleration would generate larger error.
Error Analysis of Vehicle Speed Measurement Method Based on Video in Crash Analysis
18th COTA International Conference of Transportation Professionals ; 2018 ; Beijing, China
CICTP 2018 ; 1263-1272
2018-07-02
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Vehicle speed measurement system based on intelligent video analysis
Europäisches Patentamt | 2022
|British Library Conference Proceedings | 1997
|Practical application of vehicle speed determination from crash measurement
Kraftfahrwesen | 1987
|Target-vehicle modeling in crash analysis
Tema Archiv | 2001
|