Abstract Traffic surveillance cameras are the eyes of the Intelligent Transportation Systems (ITS). However, they are currently isolated and can only extract information from each of their fixed views. To track vehicles across multiple cameras and help public agencies collect link travel time and speed information, an Edge-empowered Cooperative Multi-camera Sensing (ECoMS) System is proposed. ECoMS system presents a novel algorithmic and edge-server cooperative system construct to push edge computing and multi-camera re-identification workflow serving for traffic sensing based on Internet of Things (IoT) architecture. On the algorithm side, ECoMS system proposes a featherlight edge-based computer vision framework for vehicle detection, tracking, and features selection process in a real-time manner. Then, by only sending the objects’ representations to the server, the high-bandwidth data transmission and the heavy post-processing system can be abandoned. Furthermore, a hierarchical clip-based deep vehicle re-identification framework is proposed and integrated into the ECoMS system, and significantly outperforms other state-of-the-art methods by 4%–8% on Rank-1 accuracy. Finally, to balance the accuracy level of different camera pairs, a collaborative cross-camera traffic information estimation framework based on kernel density estimation with kernel smoother is implemented, which can get the precise link and region traffic information together with distributions (less than 1.01 KL distance). By maximizing the cooperation of the computational resources, orchestrating the data transmission and integrating road network graph features in the system, the ECoMS can precisely model the network-scale traffic information in a flexible, cost-effective, and scalable workflow. To the author’s best knowledge, ECoMS is the first multi-camera vehicle tracking and traffic monitoring system based on cooperative IoT architecture.

    Highlights An edge-server cooperative IoT workflow for multi-camera vehicle tracking and traffic perception is presented, extensively evaluated and real implemented for extracting traffic information for multiple public agencies. A multi-task deep feature extraction workflow is proposed and optimized for the edge devices, including vehicle detection, tracking and objects’ representation selection. To match the edge inputs precisely, a novel clip-based deep vehicle Re-ID model, with the hierarchical feature extraction and fusion mechanism is proposed and integrated into the cooperative perception workflow. Besides, to extract the accurate information from various camera pairs with differing Re-ID accuracies, a precision-aware kernel density estimator is integrated in the proposed system instead of using brute-force sampling approach. The cost of the proposed cooperative multi-caemra perception system is significantly reduced on the server hardware, including the GPUs (only with 25% cost) and data storage (with less than 10% of original data volume), together with the communication bandwidth.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Cooperative multi-camera vehicle tracking and traffic surveillance with edge artificial intelligence and representation learning


    Beteiligte:
    Yang, Hao (Frank) (Autor:in) / Cai, Jiarui (Autor:in) / Liu, Chenxi (Autor:in) / Ke, Ruimin (Autor:in) / Wang, Yinhai (Autor:in)


    Erscheinungsdatum :

    2022-12-04




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Multi-camera detection and multi-target tracking. Traffic surveillance applications

    Reulke, R. / Bauer, S. / Döring, T. et al. | Tema Archiv | 2008



    Self-Calibration of Traffic Surveillance Camera using Motion Tracking

    Thi, Tuan Hue / Lu, Sijun / Zhang, Jian | IEEE | 2008


    Computer Vision-Based Intelligent Traffic Surveillance: Multi-Vehicle Tracking and Detection

    Husein, Amir Mahmud / Noflianhar Lubis, Kevi / Salim Sidabutar, Daniel et al. | BASE | 2024

    Freier Zugriff

    Robust Vehicle Detection and Tracking for Traffic Surveillance

    Jung, Y.-K. / Ho, Y.-S. | British Library Conference Proceedings | 1999