Efficient maritime surveillance is necessary for navigation, which usually uses cameras to capture the vessels. However, the information in camera-based data is limited. In this work, we propose a novel intelligent maritime surveillance framework driven by the fusion of camera-based vessel detection and Automatic Identification System (AIS) data. Firstly, we employ a vessel detection network to get the relative positions of the vessels from the calibrated camera-based data. Meanwhile, we design a series of filters based on data completeness, detection range, and vessel course to exclude the invalid AIS data. In the end, we propose a data fusion module based on estimating the time when the vessel arrive at the specific position. According to the experiment on our collected dataset, the proposed framework performs competitively in diversified scenes. The mean absolute distance deviation of the estimation is less than 30 meters, and the accuracy of data fusion is 81.423%.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Intelligent Maritime Surveillance Framework Driven by Fusion of Camera-Based Vessel Detection and AIS Data


    Beteiligte:
    Qu, Jingxiang (Autor:in) / Guo, Yu (Autor:in) / Lu, Yuxu (Autor:in) / Zhu, Fenghua (Autor:in) / Huan, Yingchun (Autor:in) / Liu, Ryan Wen (Autor:in)


    Erscheinungsdatum :

    08.10.2022


    Format / Umfang :

    3102752 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Data fusion architecture for Maritime Surveillance

    Gad, A. / Farooq, M. | IEEE | 2002


    Data Fusion Architecture for Maritime Surveillance

    Gad, A. / Farooq, M. / International Society of Information Fusion et al. | British Library Conference Proceedings | 2002




    Multi-Task Learning-Enabled Automatic Vessel Draft Reading for Intelligent Maritime Surveillance

    Qu, Jingxiang / Liu, Ryan Wen / Zhao, Chenjie et al. | IEEE | 2024