Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images

The freshwater resource is invaluable and indispensable for any nation like the Republic of Korea. Recently, deep learning (DL), AI models have become more popular and applied frequently for surface water studies. The Segment Anything Model (SAM) has been developing sharply and takes an adaptable ap...

Full description

Saved in:
Bibliographic Details
Main Authors: Nguyen Hong Quang, Hanna Lee, Eui-Myoung Kim, Gihong Kim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10795130/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850023230558961664
author Nguyen Hong Quang
Hanna Lee
Eui-Myoung Kim
Gihong Kim
author_facet Nguyen Hong Quang
Hanna Lee
Eui-Myoung Kim
Gihong Kim
author_sort Nguyen Hong Quang
collection DOAJ
description The freshwater resource is invaluable and indispensable for any nation like the Republic of Korea. Recently, deep learning (DL), AI models have become more popular and applied frequently for surface water studies. The Segment Anything Model (SAM) has been developing sharply and takes an adaptable approach with the ability to perform zero-shot inference. Although, pre-trained SAM was trained with millions of images (a billion masks), applying it to remote sensing data reveals limitations of inaccurate results and unlabeled classes, particularly for the more complex and noise data from Synthetic-aperture Radar Images. Hence, we fine-tune the SAM model and other popular CNN models of YOLOv8, U-net(ResNet50), and DeepLab(ResNet50, EfficientNet) for lake semantic segmentation using multi-SAR RS datasets of Kompsat-5, ALOS-2, Sentinel-1, and a combination of the three datasets (data link) for model result comparisons. This study&#x2019;s accuracy assessment showed the SAM was the most precise model (accuracy overall <inline-formula> <tex-math notation="LaTeX">$\approx ~0.95$ </tex-math></inline-formula>) followed by the DeepLab(ResNet50), YOLOv8, U-net(ResNet50), and DeepLab(EfficientNet) model. Almost all models segmented highly accurate lake areas fitted well with ground-truth masks, nevertheless, the SAM (code link) presented the most well-performance model. The YOLO deals well with larger datasets and requires deeper trains to gain higher accuracy outputs. In addition, this research investigated the responses of each DL model to the SAR dataset proving a need for fine-tuning model results on SAR RS for better lake segmentations.
format Article
id doaj-art-869f7e9a40a84dfc8186d74bddee636d
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-869f7e9a40a84dfc8186d74bddee636d2025-08-20T03:01:27ZengIEEEIEEE Access2169-35362025-01-01131727175010.1109/ACCESS.2024.351651910795130Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar ImagesNguyen Hong Quang0https://orcid.org/0000-0001-7657-624XHanna Lee1https://orcid.org/0000-0001-7862-8306Eui-Myoung Kim2https://orcid.org/0000-0002-2835-0526Gihong Kim3https://orcid.org/0000-0003-3280-5340Institute for Smart Infrastructure, Gangneung-Wonju National University, Gangneung-si, Gangwon-do, South KoreaInstitute for Smart Infrastructure, Gangneung-Wonju National University, Gangneung-si, Gangwon-do, South KoreaDepartment of Drone and GIS Engineering, Namseoul University, Cheonan, South KoreaDepartment of Civil and Environmental Engineering, Gangneung-Wonju National University, Gangneung-si, Gangwon-do, South KoreaThe freshwater resource is invaluable and indispensable for any nation like the Republic of Korea. Recently, deep learning (DL), AI models have become more popular and applied frequently for surface water studies. The Segment Anything Model (SAM) has been developing sharply and takes an adaptable approach with the ability to perform zero-shot inference. Although, pre-trained SAM was trained with millions of images (a billion masks), applying it to remote sensing data reveals limitations of inaccurate results and unlabeled classes, particularly for the more complex and noise data from Synthetic-aperture Radar Images. Hence, we fine-tune the SAM model and other popular CNN models of YOLOv8, U-net(ResNet50), and DeepLab(ResNet50, EfficientNet) for lake semantic segmentation using multi-SAR RS datasets of Kompsat-5, ALOS-2, Sentinel-1, and a combination of the three datasets (data link) for model result comparisons. This study&#x2019;s accuracy assessment showed the SAM was the most precise model (accuracy overall <inline-formula> <tex-math notation="LaTeX">$\approx ~0.95$ </tex-math></inline-formula>) followed by the DeepLab(ResNet50), YOLOv8, U-net(ResNet50), and DeepLab(EfficientNet) model. Almost all models segmented highly accurate lake areas fitted well with ground-truth masks, nevertheless, the SAM (code link) presented the most well-performance model. The YOLO deals well with larger datasets and requires deeper trains to gain higher accuracy outputs. In addition, this research investigated the responses of each DL model to the SAR dataset proving a need for fine-tuning model results on SAR RS for better lake segmentations.https://ieeexplore.ieee.org/document/10795130/Fine-tunningRepublic of KoreareservoirSAMspace-borne SAR
spellingShingle Nguyen Hong Quang
Hanna Lee
Eui-Myoung Kim
Gihong Kim
Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
IEEE Access
Fine-tunning
Republic of Korea
reservoir
SAM
space-borne SAR
title Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
title_full Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
title_fullStr Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
title_full_unstemmed Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
title_short Fine-Tunned Segment Anything Model (SAM) for Reservoir Extractions Compared With Popular CNNs: An Experiment for Space-Borne Synthetic-Aperture Radar Images
title_sort fine tunned segment anything model sam for reservoir extractions compared with popular cnns an experiment for space borne synthetic aperture radar images
topic Fine-tunning
Republic of Korea
reservoir
SAM
space-borne SAR
url https://ieeexplore.ieee.org/document/10795130/
work_keys_str_mv AT nguyenhongquang finetunnedsegmentanythingmodelsamforreservoirextractionscomparedwithpopularcnnsanexperimentforspacebornesyntheticapertureradarimages
AT hannalee finetunnedsegmentanythingmodelsamforreservoirextractionscomparedwithpopularcnnsanexperimentforspacebornesyntheticapertureradarimages
AT euimyoungkim finetunnedsegmentanythingmodelsamforreservoirextractionscomparedwithpopularcnnsanexperimentforspacebornesyntheticapertureradarimages
AT gihongkim finetunnedsegmentanythingmodelsamforreservoirextractionscomparedwithpopularcnnsanexperimentforspacebornesyntheticapertureradarimages