Since its inception, the maritime transport industry has occupied a prominent position among the five modes of transport. The study of shipping risk in sea and waterway area, especially the decision-making analysis of ship collision risk, is of great significance for shipping safety engineering, reducing Marine accidents, pollution prevention, and emergency planning. Existing risk assessment models require experts to interpret the model results, which is time-consuming and laborious. In order to solve this problem, a new method of ship encounter scene recognition is proposed according to the collision risk. The complex Marine environment has high requirements for the target detection algorithm, which requires not only a high accuracy rate, but also a high real-time performance, which also requires the algorithm to have good robustness. Aiming at the shortcomings of the traditional target detection algorithm with low detection accuracy and poor robustness, this paper proposes a detection method based on the improved YOLOv5 algorithm model. This method uses convolutional neural network in deep learning to process the images of two or more ships, and makes use of the powerful fitting ability of convolutional neural network. More accurate use of image features to identify boats on the water.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Shipping Risk Assessment Ship Inspection Method Based on Improved YOLOv5


    Weitere Titelangaben:

    Lect. Notes Electrical Eng.


    Beteiligte:
    Yadav, Sanjay (Herausgeber:in) / Arya, Yogendra (Herausgeber:in) / Pandey, Shailesh M. (Herausgeber:in) / Gherabi, Noredine (Herausgeber:in) / Karras, Dimitrios A. (Herausgeber:in) / Wang, Ziqi (Autor:in) / Li, Haibin (Autor:in) / Sun, Mingzhu (Autor:in) / Zhu, Louyu (Autor:in)

    Kongress:

    International Conference on Artificial Intelligence, Robotics, and Communication ; 2023 ; Xiamen, China December 22, 2023 - December 24, 2023



    Erscheinungsdatum :

    19.06.2024


    Format / Umfang :

    9 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    STD-Yolov5: a ship-type detection model based on improved Yolov5

    Ning, Yue / Zhao, Lining / Zhang, Can et al. | Taylor & Francis Verlag | 2024


    Ship-YOLOv5s: improved YOLOv5 ship target detection based on attention mechanism

    Zhang, Peng / Sun, Ze / Dong, Junwei et al. | SPIE | 2025


    Research on ship detection technology based on improved YOLOv5

    Huan, Yutai / Chen, Lin / Liu, Bin et al. | IEEE | 2023