Underwater object detection and recognition rely heavily on the quality of sonar images, which can be significantly impacted by noise. This paper proposes a multi-scale denoising method for forward-looking sonar images. The proposed method firstly decomposes sonar image into three layers containing edge, texture and content information by the Laplacian pyramid model. Next, the adaptive median filter is applied to the edge layer while the modified adaptive local filter is applied to the texture layer and the content layer is completely retained. Finally, all processed layers are synthesized by Laplacian pyramid inverse transform. Experimental and comparative results on real and simulated sonar images demonstrate that the proposed method effectively reduces speckle noise and improves image quality.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A Multi-Scale Denoising Method for Forward-Looking Sonar Image


    Beteiligte:
    Wang, Zhisen (Autor:in) / Chen, Deshan (Autor:in) / Li, Zhuoyi (Autor:in) / Wu, Bing (Autor:in) / Zhao, Jinglong (Autor:in)


    Erscheinungsdatum :

    04.08.2023


    Format / Umfang :

    784691 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Underwater vehicle path planning using a multi-beam forward looking sonar

    Petillot, Y. / Tena Ruiz, I. / Lane, D.M. et al. | Tema Archiv | 1998


    Water spraying type underwater robot based on multi-beam forward-looking sonar

    SHI JIAN / LI SHIWEN / LUO YU et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    A Real-Time Pipeline Tracking Using a Forward-Looking Sonar

    Tipsuwan, Yodyium / Kasetkasem, Teerasit / Hoonsuwan, Phakhachon et al. | British Library Conference Proceedings | 2019


    SAFE NAVIGATION IN HAZARDOUS AREAS: FORWARD-LOOKING SONAR FOR SUBMARINES

    Meister, M. / Zindel, G. | British Library Online Contents | 2012