ORIGINAL RESEARCH article
Front. Remote Sens.
Sec. Image Analysis and Classification
Volume 6 - 2025 | doi: 10.3389/frsen.2025.1578841
This article is part of the Research TopicRemote Sensing Applications in Oceanography with Deep LearningView all 18 articles
A Deep-Learning Framework to Detect Green Tide from MODIS Images
Provisionally accepted
- 1Shanghai Ocean University, Shanghai, China
- 2School of Surveying and Geo-Informatics, Tongji University, Shanghai, Shanghai Municipality, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
MODIS images, renowned for their broad spatial coverage, high temporal resolution, and extensive historical archives, are widely employed for long-term monitoring of Ulva prolifera (U. prolifera) blooms. However, their relatively low spatial resolution can introduce inaccuracies, potentially hindering precise delineation of bloom extents and the formulation of effective management strategies. Our study introduces WaveNet, a model integrating VGG16 and Bidirectional Feature Pyramid Network (BiFPN), coupled with a Convolutional Block Attention Module (CBAM). When applied to MODIS images, WaveNet demonstrates superior performance in long-term monitoring of U. prolifera on the sea surface compared to traditional methods. Specifically, it achieves an accuracy of 97.14% and an F1 score of 93.26%, representing a significant improvement over existing approaches. This advancement highlights its enhanced capability in precise spatial recognition and classification, addressing limitations in prior techniques and setting a new benchmark for green tide monitoring. Utilizing this approach, we analyzed the U. prolifera blooms in the Yellow Sea of China from 2018 to 2024, offering actionable insights for early prevention and targeted management of green tides. By significantly enhancing prediction accuracy and improving traditional monitoring practices, this method supports the development of more effective mitigation strategies to reduce the environmental impact of U. prolifera blooms.
Keywords: Deep learning model, green tide detection, MODIS, satellite remote sensing, Yellow Sea
Received: 18 Feb 2025; Accepted: 28 Jul 2025.
Copyright: © 2025 Zhu, Xu, Zhang, Liu, Cao, Liu and Li. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Yuelin Xu, Shanghai Ocean University, Shanghai, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.