In order to solve the problems of unbalanced sampling of track points in the track point distance-based ship track clustering algorithm, the abnormality of individual track points affecting the clustering effect, and the difficulty of accurately describing the spatial characteristics of ship tracks with the latitude and longitude data of track points, a Kmeans ship track clustering algorithm based on the similarity of track image features is proposed. The algorithm defines a similarity measure based on trajectory image features. The method converts ship trajectory latitude and longitude time series data into ship trajectory image data, extracts trajectory image features using ResNet-50, and uses the Euclidean distance between trajectory image features as a method to measure the similarity of ship trajectories. Using the similarity measure based on the trajectory image features, the ship trajectories are clustered by the Kmeans algorithm. Experimental results show that the proposed algorithm improves the accuracy by 10% over the traditional DTW-based K-centroid clustering algorithm, and can cluster a large number of complex ship trajectories, and the clustering results are consistent with the actual traffic flow.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    K-means ship trajectory clustering algorithm based on trajectory image similarity


    Contributors:
    Yue, Yang (editor) / Xu, Shuwen (editor) / Shi, Qi (author) / Fan, Yaqiong (author) / Zhang, Danpu (author) / Guo, Huiming (author) / Liu, Jia (author)

    Conference:

    Third International Conference on Digital Signal and Computer Communications (DSCC 2023) ; 2023 ; Xi'an, China


    Published in:

    Proc. SPIE ; 12716


    Publication date :

    2023-07-28





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Ship route prediction method based on ship trajectory clustering

    LIANG SHAN / SHANG YONGKANG / WANG DEJUN et al. | European Patent Office | 2021

    Free access

    Similarity Based Vehicle Trajectory Clustering and Anomaly Detection

    Fu, Z. / Hu, W. / Tan, T. | British Library Conference Proceedings | 2005


    Similarity based vehicle trajectory clustering and anomaly detection

    Zhouyu Fu, / Weiming Hu, / Tieniu Tan, | IEEE | 2005


    A novel method for ship trajectory clustering

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

    Free access