Ship trajectory clustering is one of the main methods for mining ship feature trajectories based on AIS data. However, there are two main problems in trajectory clustering: First, the clustering algorithm itself has the problems of difficult to determine the parameters and poor noise recognition ability; second, the trajectory similarity metric, most of the measurements are only similarity metrics for the ship’s position, and do not take into account the other dimensional information of the ship’s trajectory. In order to solve these problems, this paper proposes a fast adaptive density clustering method for ship trajectories, which integrally considers multiple attributes of ship position, heading and speed to construct similarity metrics between ship trajectories; introduces Silhouette Coefficient (SC) and Davies-Bouldin Index (DBI) The DBTCAN algorithm is constructed to evaluate the comprehensive CMI index, which in turn realizes the adaptive selection of clustering parameters. An example study was conducted using AIS data of real waters, and the results show that the method can adaptively cluster ship trajectories to match the traffic situation of real waters.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Adaptive Density Ship Trajectory Clustering Based on AIS data


    Weitere Titelangaben:

    Advances in Engineering res


    Beteiligte:
    Zhao, Gaofeng (Herausgeber:in) / Satyanaga, Alfrendo (Herausgeber:in) / Ramani, Sujatha Evangelin (Herausgeber:in) / Abdel Raheem, Shehata E. (Herausgeber:in) / Wu, Changsheng (Autor:in) / He, Shuchang (Autor:in)

    Kongress:

    International Symposium on Traffic Transportation and Civil Architecture ; 2024 ; Tianjin, China June 21, 2024 - June 23, 2024



    Erscheinungsdatum :

    24.09.2024


    Format / Umfang :

    10 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Ship route prediction method based on ship trajectory clustering

    LIANG SHAN / SHANG YONGKANG / WANG DEJUN et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Adaptive mining algorithm of ship typical trajectory based on improved spectral clustering

    Liu, Chang / Zhang, Shize / Ling, Yuan et al. | British Library Conference Proceedings | 2022




    A novel method for ship trajectory clustering

    Helong Shen / Huang Tang / Yong Yin | DOAJ | 2022

    Freier Zugriff