Due to the increasing demand for multi-vehicle speed measurement on urban roads, as well as the cost and environmental constraints of using laser and ultrasonic speed measurement methods, we propose an accurate multi-vehicle speed measurement method based on urban viaduct monitoring video without additional equipment and labels. This paper first adopts an improved multi-object tracking (MOT) algorithm to achieve high-precision tracking of multi-vehicles in urban viaduct monitoring videos. Then, the mapping of the image space to the actual space is realized by using the novel space transformation method designed independently to complete the measurement of the actual vehicle speed. At the same time, the Kalman Filter and Polynomial Fitting are employed to eliminate the errors in the speed measurement process and further improve the measurement accuracy. Finally, the effectiveness of the method was demonstrated by its application to an urban viaduct in Shanghai, China. The velocity measurement error in straight and curved scenes can be controlled within 0.5km/h, and 1km/h, respectively. The final velocity curve can accurately reflect changes in vehicle speed.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Multi-vehicle Speed Measurement Based on Monitoring Video Using Improved Multi-object Tracking Algorithm


    Contributors:
    Gao, Xinwen (author) / Gan, Zhiyuan (author) / Shen, Zhuo (author) / Teng, Li (author)


    Publication date :

    2023-07-27


    Size :

    8116909 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Automatic multi-vehicle tracking using video cameras: An improved CAMShift approach

    Xia, Jingxin / Rao, Wenming / Huang, Wei et al. | Springer Verlag | 2013


    Automatic multi-vehicle tracking using video cameras: An improved CAMShift approach

    Xia, Jingxin / Rao, Wenming / Huang, Wei et al. | Online Contents | 2013



    Multi-Object Tracking in Video using a Trisection Paradigm

    Al-Hamadi, A. / Niese, R. / Michealis, B. | British Library Online Contents | 2005