Next Article in Journal
Seasonal and Interannual Variations in Hydrological Dynamics of the Amazon Basin: Insights from Geodetic Observations
Previous Article in Journal
Improved Photosynthetic Accumulation Models for Biomass Estimation of Soybean and Cotton Using Vegetation Indices and Canopy Height
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

Deep Learning Small Water Body Mapping by Transfer Learning from Sentinel-2 to PlanetScope

1
Key Laboratory for Environment and Disaster Monitoring and Evaluation, Hubei, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences, Wuhan 430077, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(15), 2738; https://doi.org/10.3390/rs17152738 (registering DOI)
Submission received: 15 June 2025 / Revised: 24 July 2025 / Accepted: 6 August 2025 / Published: 7 August 2025
(This article belongs to the Section Remote Sensing Image Processing)

Abstract

Small water bodies are widely spread and play crucial roles in supporting regional agricultural and aquaculture activities. PlanetScope imagery has a high resolution (3 m) with daily global coverage and has obviously enhanced small water body mapping. Recent studies have demonstrated the effectiveness of deep learning for mapping small water bodies using PlanetScope; however, a persistent challenge remains in the scarcity of high-quality, manually annotated water masks used for model training, which limits the generalization capability of data-driven deep learning models. In this study, we propose a transfer learning framework that leverages Sentinel-2 data to improve PlanetScope-based small water body mapping, capitalizing on the spectral interoperability between PlanetScope and Sentinel-2 bands and the abundance of open-source Sentinel-2 water masks. Eight state-of-the-art segmentation models have been explored. Additionally, this paper presents the first assessment of the VMamba model for small water body mapping, building on its demonstrated success in segmentation tasks. The models were pre-trained using Sentinel-2-derived water masks and subsequently fine-tuned with a limited set (1292 image patches, 256 × 256 pixels in each patch) of manually annotated PlanetScope labels. Experiments were conducted using 5648 image patches and two areas of 9636 km2 and 2745 km2, respectively. Among the evaluated methods, VMamba achieved higher accuracy compared with both CNN- and Transformer-based models. This study highlights the efficacy of combining global Sentinel-2 datasets for pre-training with localized fine-tuning, which not only enhances mapping accuracy but also reduces reliance on labor-intensive manual annotation in regional small water body mapping.
Keywords: small water body; VMamba; transfer learning; fine-tuning small water body; VMamba; transfer learning; fine-tuning

Share and Cite

MDPI and ACS Style

Li, Y.; Zhou, P.; Wang, Y.; Li, X.; Zhang, Y.; Li, X. Deep Learning Small Water Body Mapping by Transfer Learning from Sentinel-2 to PlanetScope. Remote Sens. 2025, 17, 2738. https://doi.org/10.3390/rs17152738

AMA Style

Li Y, Zhou P, Wang Y, Li X, Zhang Y, Li X. Deep Learning Small Water Body Mapping by Transfer Learning from Sentinel-2 to PlanetScope. Remote Sensing. 2025; 17(15):2738. https://doi.org/10.3390/rs17152738

Chicago/Turabian Style

Li, Yuyang, Pu Zhou, Yalan Wang, Xiang Li, Yihang Zhang, and Xiaodong Li. 2025. "Deep Learning Small Water Body Mapping by Transfer Learning from Sentinel-2 to PlanetScope" Remote Sensing 17, no. 15: 2738. https://doi.org/10.3390/rs17152738

APA Style

Li, Y., Zhou, P., Wang, Y., Li, X., Zhang, Y., & Li, X. (2025). Deep Learning Small Water Body Mapping by Transfer Learning from Sentinel-2 to PlanetScope. Remote Sensing, 17(15), 2738. https://doi.org/10.3390/rs17152738

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop