Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM)
Abstract Semantic segmentation is a critical part of observation-driven research in glaciology. Using remote sensing to quantify how features change (e.g. glacier termini, supraglacial lakes, icebergs, crevasses) is particularly important in polar regions, where glaciological features may be spatial...
Published in: | Journal of Glaciology |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Cambridge University Press (CUP)
2023
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1017/jog.2023.95 https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0022143023000953 |
id |
crcambridgeupr:10.1017/jog.2023.95 |
---|---|
record_format |
openpolar |
spelling |
crcambridgeupr:10.1017/jog.2023.95 2024-09-15T18:15:37+00:00 Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) Shankar, Siddharth Stearns, Leigh A. van der Veen, C. J. 2023 http://dx.doi.org/10.1017/jog.2023.95 https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0022143023000953 en eng Cambridge University Press (CUP) http://creativecommons.org/licenses/by/4.0/ Journal of Glaciology page 1-10 ISSN 0022-1430 1727-5652 journal-article 2023 crcambridgeupr https://doi.org/10.1017/jog.2023.95 2024-07-10T04:04:42Z Abstract Semantic segmentation is a critical part of observation-driven research in glaciology. Using remote sensing to quantify how features change (e.g. glacier termini, supraglacial lakes, icebergs, crevasses) is particularly important in polar regions, where glaciological features may be spatially small but reflect important shifts in boundary conditions. In this study, we assess the utility of the Segment Anything Model (SAM), released by Meta AI Research, for cryosphere research. SAM is a foundational AI model that generates segmentation masks without additional training data. This is highly beneficial in polar science because pre-existing training data rarely exist. Widely-used conventional deep learning models such as UNet require tens of thousands of training labels to perform effectively. We show that the Segment Anything Model performs well for different features (icebergs, glacier termini, supra-glacial lakes, crevasses), in different environmental settings (open water, mélange, and sea ice), with different sensors (Sentinel-1, Sentinel-2, Planet, timelapse photographs) and different spatial resolutions. Due to the performance, versatility, and cross-platform adaptability of SAM, we conclude that it is a powerful and robust model for cryosphere research. Article in Journal/Newspaper Journal of Glaciology Sea ice Cambridge University Press Journal of Glaciology 1 10 |
institution |
Open Polar |
collection |
Cambridge University Press |
op_collection_id |
crcambridgeupr |
language |
English |
description |
Abstract Semantic segmentation is a critical part of observation-driven research in glaciology. Using remote sensing to quantify how features change (e.g. glacier termini, supraglacial lakes, icebergs, crevasses) is particularly important in polar regions, where glaciological features may be spatially small but reflect important shifts in boundary conditions. In this study, we assess the utility of the Segment Anything Model (SAM), released by Meta AI Research, for cryosphere research. SAM is a foundational AI model that generates segmentation masks without additional training data. This is highly beneficial in polar science because pre-existing training data rarely exist. Widely-used conventional deep learning models such as UNet require tens of thousands of training labels to perform effectively. We show that the Segment Anything Model performs well for different features (icebergs, glacier termini, supra-glacial lakes, crevasses), in different environmental settings (open water, mélange, and sea ice), with different sensors (Sentinel-1, Sentinel-2, Planet, timelapse photographs) and different spatial resolutions. Due to the performance, versatility, and cross-platform adaptability of SAM, we conclude that it is a powerful and robust model for cryosphere research. |
format |
Article in Journal/Newspaper |
author |
Shankar, Siddharth Stearns, Leigh A. van der Veen, C. J. |
spellingShingle |
Shankar, Siddharth Stearns, Leigh A. van der Veen, C. J. Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
author_facet |
Shankar, Siddharth Stearns, Leigh A. van der Veen, C. J. |
author_sort |
Shankar, Siddharth |
title |
Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
title_short |
Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
title_full |
Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
title_fullStr |
Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
title_full_unstemmed |
Semantic segmentation of glaciological features across multiple remote sensing platforms with the Segment Anything Model (SAM) |
title_sort |
semantic segmentation of glaciological features across multiple remote sensing platforms with the segment anything model (sam) |
publisher |
Cambridge University Press (CUP) |
publishDate |
2023 |
url |
http://dx.doi.org/10.1017/jog.2023.95 https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0022143023000953 |
genre |
Journal of Glaciology Sea ice |
genre_facet |
Journal of Glaciology Sea ice |
op_source |
Journal of Glaciology page 1-10 ISSN 0022-1430 1727-5652 |
op_rights |
http://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.1017/jog.2023.95 |
container_title |
Journal of Glaciology |
container_start_page |
1 |
op_container_end_page |
10 |
_version_ |
1810453474115059712 |