As a coastal state, China's maritime trade and maritime security are crucial, and efficient and accurate detection and recognition of ships is an essential part of this. However, currently, deep learning-based ship target detection methods suffer from scarce datasets and imbalanced samples, leading to weak generalization capabilities of many detection models that are unable to adapt to a variety of environments. In this article, a new large marine ship dataset was established, and the AutoAugment data processing method was used to process ship image information. The YOLOv3 algorithm was utilized for ship target detection, improving the robustness and prediction accuracy of ship detection under complex coastal conditions, achieving high-quality detection of ships at sea.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Ship target detection based on YOLOv3 algorithm


    Contributors:
    Wang, Yao (author) / Wang, Dongzhuo (author)


    Publication date :

    2023-08-26


    Size :

    1039068 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Ship Target Detection Algorithm Based on Improved YOLOv3 for Maritime Image

    Dehai Chen / Shiru Sun / Zhijun Lei et al. | DOAJ | 2021

    Free access

    SAR SHIP DETECTION BASED ON IMPROVED YOLOV3

    Chen, D. / Ju, Y. | TIBKAT | 2021



    Highway curve bidirectional early warning system based on YOLOV3 target detection algorithm

    DUAN LI / SHEN ZHENWU | European Patent Office | 2022

    Free access