This work presents a method for detecting and tracking vehicles in traffic videos collected from urban highways. The method efficiently handles numerous vehicle targets with low computational complexity, making it ideal for real-time driver assisting systems. The method uses segmentation to create a quick bounding box to detect moving vehicles, while a particle filter is used to track these vehicles. The Hungarian matching algorithm is later used to solve the association of tracked vehicles in each video frame. The proposed method and an established vehicle tracking method, Camshift Meanshift tracking with Kalman filtering (CMT-KF), were compared and evaluated on the collected traffic videos. The proposed method resulted in a 52% reduction in tracking failures.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Hungarian-Particle Filtering Based Segmentation for On-Road Visual Vehicle Detection and Tracking




    Publication date :

    2022-03-07


    Size :

    1052894 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English