Pedestrian target tracking based on the UAV platform can be widely used in traffic control, field search, and military reconnaissance. It is an important research task of computer vision and intelligent cruise. Aiming at the limitations of the UAV surveillance system in moving pedestrian target tracking, such as background change, pedestrian deformation, occlusion interference, and lack of real-time performance, the dual Kalman filter is used to improve the traditional TLD tracking algorithm, the proposed method can accelerate the correction of the predicted detection area, reduce the disturbance of the environment background and the target deformation to the pedestrian tracking accuracy, and reduce the detection time by using the adaptive adjustment method of the detection area to offset the time cost caused by double Kalman filtering, to improve the Algorithm’s real-time performance. The test results show that the proposed method has high accuracy, stability, and real-time performance in pedestrian target tracking based on the UAV platform.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Pedestrian target tracking of UAV platform based on improved TLD algorithm


    Beteiligte:
    Zhou, Bin (Autor:in) / Jin, Ali (Autor:in) / Zhang, Yu (Autor:in) / Zhang, Yudi (Autor:in) / Zhao, Mingdong (Autor:in)

    Kongress:

    AOPC 2022: Optical Sensing, Imaging, and Display Technology ; 2022 ; Beijing,China


    Erschienen in:

    Proc. SPIE ; 12557


    Erscheinungsdatum :

    23.01.2023





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Visual Pedestrian Tracking from a UAV Platform

    Zhang, Li / Zhang, Zenghui / Xiong, Huilin | IEEE | 2017


    Pedestrian Detection Algorithm Based on Improved YOLOv4

    Zheng, Ziheng / Ni, Chenhui / Zeng, Guolei | IEEE | 2023



    Research on Vision-based pedestrian detection and tracking algorithm

    Sun, Yao / Yan, Yongjie / Zhao, Jie et al. | British Library Conference Proceedings | 2022


    Detection Algorithm Based on Improved YOLOv4 for Pedestrian

    Kong, Wen / Qiao, Yidan / Wei, Ziqi | IEEE | 2022