Detection of ships and precise recognition within expansive remote sensing images constitutes a pivotal research avenue within the domain of remote sensing image analysis. The primary objective of this study is to enhance ship detection accuracy in satellite imagery using a novel multi-loss deep learning architecture. The following paper presents a novel multi-loss deep learning model for detecting ships in satellite imagery. The proposed model uses focal loss, Dice loss functions and binary cross-entropy to improve detection performance and has achieved an accuracy of 99.40% on a large dataset of satellite images. The model outperforms existing methods and provides an innovative solution to ship detection in satellite imagery. Its performance was validated through comprehensive evaluation and comparison with other state-of-the-art models, highlighting its effectiveness and computational efficiency. The proposed model's potential for enhancing ocean monitoring and security is significant, with applications in maritime security, illegal activity monitoring, and shipping route optimization. This research demonstrates the importance of deep learning-based methods in maritime ship detection from satellite imagery and emphasizes the need for continued innovation in this field.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Maritime Vessel Segmentation in Satellite Imagery Using UNET Architecture and Multiloss Optimization


    Weitere Titelangaben:

    Communic.Comp.Inf.Science


    Beteiligte:
    Satheeskumaran, S. (Herausgeber:in) / Zhang, Yudong (Herausgeber:in) / Balas, Valentina Emilia (Herausgeber:in) / Hong, Tzung-pei (Herausgeber:in) / Pelusi, Danilo (Herausgeber:in) / Ghadekar, Premanand (Autor:in) / Deshpande, Mihir (Autor:in) / Gharpure, Adwait (Autor:in) / Gokhale, Vedant (Autor:in) / Gore, Aayush (Autor:in)

    Kongress:

    International Conference on Intelligent Computing for Sustainable Development ; 2023 ; Hyderabad, India August 25, 2023 - August 26, 2023



    Erscheinungsdatum :

    24.05.2024


    Format / Umfang :

    14 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Automatic Maritime Object Detection Using Satellite imagery

    Bereta, Konstantina / Zissis, Dimitris / Grasso, Raffaele | IEEE | 2020


    VESSEL MONITORING SYSTEM AND VESSEL MONITORING METHOD USING SATELLITE IMAGERY

    KIM MIN SIK / PARK SEONG WOOK | Europäisches Patentamt | 2021

    Freier Zugriff


    SPOT IMAGE/SARCOM: Satellite Imagery to Reinforce Maritime Surveillance

    Baillarin-Ribbes, F. / Chikhi, M. / Campenon, P. | TIBKAT | 2005


    Moth image segmentation based on improved Unet

    Sun, Qilin / Zhang, Ruirui / Chen, Liping et al. | British Library Conference Proceedings | 2022