Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 10, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

999

MGPNet: Rethinking multi-scale features and global attention with pre-trained model for SAR oil spill detection.

Shaokang Dong1, Jiangfan Feng2

  • 1Chongqing Engineering Research Center of Spatial Big Data Intelligent Technology, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China; School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China.

Marine Pollution Bulletin
|November 23, 2025
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A non-invasive framework for automated fish monitoring and morphological trait extraction: Implications for ecological monitoring and sustainable fisheries.

Marine pollution bulletin·2026
Same author

A unified and efficient training framework for open-ended non-transitive games.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

Bayesian Optimization-Enhanced Reinforcement learning for Self-adaptive and multi-objective control of wastewater treatment.

Bioresource technology·2025
Same author

SAGPNet: A shape-aware and adaptive strip self-attention guided progressive network for SAR marine oil spill detection.

Marine environmental research·2024
Same author

Coordinating Multi-Agent Reinforcement Learning via Dual Collaborative Constraints.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Multi-Task Multi-Agent Reinforcement Learning With Interaction and Task Representations.

IEEE transactions on neural networks and learning systems·2024
This summary is machine-generated.

Marine oil spill detection using Synthetic Aperture Radar (SAR) imagery is improved by MGPNet. This novel network effectively addresses limited data, look-alike confusion, and appearance variability for accurate oceanic environmental monitoring.

Area of Science:

  • Remote Sensing
  • Oceanography
  • Artificial Intelligence

Background:

  • Marine oil spills cause significant ecological damage.
  • Synthetic Aperture Radar (SAR) imagery is crucial for oil spill detection.
  • Existing SAR-based methods face challenges like limited datasets, look-alike interference, and variable target appearances, hindering accuracy.

Purpose of the Study:

  • To propose a novel network, MGPNet, for enhanced oil spill detection from SAR imagery.
  • To address the limitations of existing methods, including data scarcity and interference from look-alike substances.
  • To improve the accuracy and robustness of marine oil spill detection.

Main Methods:

  • Developed MGPNet, a novel network incorporating multi-scale features and global attention.
Keywords:
Global attentionMarine pollutionMulti-scale featureOil spill detectionPre-trained model

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

727

Related Experiment Videos

Last Updated: Jan 10, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

999
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

727
  • Utilized pre-trained models to optimize the encoder and mitigate limited training data issues.
  • Introduced MFSF and GMFSF blocks to capture multi-scale features, suppress irrelevant information, and enhance feature representation against look-alike interference.
  • Designed a novel multi-scale feature connection block to replace traditional skip connections for improved detail recovery during decoding.

    • Main Results:

    • MGPNet achieved state-of-the-art performance on three public SAR oil spill detection datasets (PALSAR, Sentinel, M4D).
    • Demonstrated significant mIoU (mean Intersection over Union) gains: 0.59% on PALSAR, 0.93% on Sentinel, and 4.24% on M4D.
    • Validated the effectiveness of MGPNet in overcoming challenges like limited data and look-alike substances.

    Conclusions:

    • MGPNet offers a robust and accurate solution for marine oil spill detection using SAR imagery.
    • The proposed network architecture effectively handles multi-scale features and global attention for improved detection.
    • MGPNet shows significant potential for practical applications in oceanic environmental monitoring and oil spill response.