Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations
Spatially dense and continuous information on avalanche occurrences is crucial for numerous safety-related applications such as avalanche warning, hazard zoning, hazard mitigation measures, forestry, risk management and numerical simulations. This information is today still collected in a non-system...
Published in: | The Cryosphere |
---|---|
Main Authors: | , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Copernicus Publications
2022
|
Subjects: | |
Online Access: | https://doi.org/10.5194/tc-16-3517-2022 https://doaj.org/article/d3d0b3d5888a42db999b3fb67624d1c7 |
id |
ftdoajarticles:oai:doaj.org/article:d3d0b3d5888a42db999b3fb67624d1c7 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:d3d0b3d5888a42db999b3fb67624d1c7 2023-05-15T18:32:27+02:00 Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations E. D. Hafner P. Barton R. C. Daudt J. D. Wegner K. Schindler Y. Bühler 2022-09-01T00:00:00Z https://doi.org/10.5194/tc-16-3517-2022 https://doaj.org/article/d3d0b3d5888a42db999b3fb67624d1c7 EN eng Copernicus Publications https://tc.copernicus.org/articles/16/3517/2022/tc-16-3517-2022.pdf https://doaj.org/toc/1994-0416 https://doaj.org/toc/1994-0424 doi:10.5194/tc-16-3517-2022 1994-0416 1994-0424 https://doaj.org/article/d3d0b3d5888a42db999b3fb67624d1c7 The Cryosphere, Vol 16, Pp 3517-3530 (2022) Environmental sciences GE1-350 Geology QE1-996.5 article 2022 ftdoajarticles https://doi.org/10.5194/tc-16-3517-2022 2022-12-30T21:09:35Z Spatially dense and continuous information on avalanche occurrences is crucial for numerous safety-related applications such as avalanche warning, hazard zoning, hazard mitigation measures, forestry, risk management and numerical simulations. This information is today still collected in a non-systematic way by observers in the field. Current research has explored the application of remote sensing technology to fill this information gap by providing spatially continuous information on avalanche occurrences over large regions. Previous investigations have confirmed the high potential of avalanche mapping from remotely sensed imagery to complement existing databases. Currently, the bottleneck for fast data provision from optical data is the time-consuming manual mapping. In our study we deploy a slightly adapted DeepLabV3+, a state-of-the-art deep learning model, to automatically identify and map avalanches in SPOT 6/7 imagery from 24 January 2018 and 16 January 2019. We relied on 24 778 manually annotated avalanche polygons split into geographically disjointed regions for training, validating and testing. Additionally, we investigate generalization ability by testing our best model configuration on SPOT 6/7 data from 6 January 2018 and comparing it to avalanches we manually annotated for that purpose. To assess the quality of the model results, we investigate the probability of detection (POD), the positive predictive value (PPV) and the F1 score. Additionally, we assessed the reproducibility of manually annotated avalanches in a small subset of our data. We achieved an average POD of 0.610, PPV of 0.668 and an F1 score of 0.625 in our test areas and found an F1 score in the same range for avalanche outlines annotated by different experts. Our model and approach are an important step towards a fast and comprehensive documentation of avalanche periods from optical satellite imagery in the future, complementing existing avalanche databases. This will have a large impact on safety-related applications, making ... Article in Journal/Newspaper The Cryosphere Directory of Open Access Journals: DOAJ Articles The Cryosphere 16 9 3517 3530 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Environmental sciences GE1-350 Geology QE1-996.5 |
spellingShingle |
Environmental sciences GE1-350 Geology QE1-996.5 E. D. Hafner P. Barton R. C. Daudt J. D. Wegner K. Schindler Y. Bühler Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
topic_facet |
Environmental sciences GE1-350 Geology QE1-996.5 |
description |
Spatially dense and continuous information on avalanche occurrences is crucial for numerous safety-related applications such as avalanche warning, hazard zoning, hazard mitigation measures, forestry, risk management and numerical simulations. This information is today still collected in a non-systematic way by observers in the field. Current research has explored the application of remote sensing technology to fill this information gap by providing spatially continuous information on avalanche occurrences over large regions. Previous investigations have confirmed the high potential of avalanche mapping from remotely sensed imagery to complement existing databases. Currently, the bottleneck for fast data provision from optical data is the time-consuming manual mapping. In our study we deploy a slightly adapted DeepLabV3+, a state-of-the-art deep learning model, to automatically identify and map avalanches in SPOT 6/7 imagery from 24 January 2018 and 16 January 2019. We relied on 24 778 manually annotated avalanche polygons split into geographically disjointed regions for training, validating and testing. Additionally, we investigate generalization ability by testing our best model configuration on SPOT 6/7 data from 6 January 2018 and comparing it to avalanches we manually annotated for that purpose. To assess the quality of the model results, we investigate the probability of detection (POD), the positive predictive value (PPV) and the F1 score. Additionally, we assessed the reproducibility of manually annotated avalanches in a small subset of our data. We achieved an average POD of 0.610, PPV of 0.668 and an F1 score of 0.625 in our test areas and found an F1 score in the same range for avalanche outlines annotated by different experts. Our model and approach are an important step towards a fast and comprehensive documentation of avalanche periods from optical satellite imagery in the future, complementing existing avalanche databases. This will have a large impact on safety-related applications, making ... |
format |
Article in Journal/Newspaper |
author |
E. D. Hafner P. Barton R. C. Daudt J. D. Wegner K. Schindler Y. Bühler |
author_facet |
E. D. Hafner P. Barton R. C. Daudt J. D. Wegner K. Schindler Y. Bühler |
author_sort |
E. D. Hafner |
title |
Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
title_short |
Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
title_full |
Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
title_fullStr |
Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
title_full_unstemmed |
Automated avalanche mapping from SPOT 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
title_sort |
automated avalanche mapping from spot 6/7 satellite imagery with deep learning: results, evaluation, potential and limitations |
publisher |
Copernicus Publications |
publishDate |
2022 |
url |
https://doi.org/10.5194/tc-16-3517-2022 https://doaj.org/article/d3d0b3d5888a42db999b3fb67624d1c7 |
genre |
The Cryosphere |
genre_facet |
The Cryosphere |
op_source |
The Cryosphere, Vol 16, Pp 3517-3530 (2022) |
op_relation |
https://tc.copernicus.org/articles/16/3517/2022/tc-16-3517-2022.pdf https://doaj.org/toc/1994-0416 https://doaj.org/toc/1994-0424 doi:10.5194/tc-16-3517-2022 1994-0416 1994-0424 https://doaj.org/article/d3d0b3d5888a42db999b3fb67624d1c7 |
op_doi |
https://doi.org/10.5194/tc-16-3517-2022 |
container_title |
The Cryosphere |
container_volume |
16 |
container_issue |
9 |
container_start_page |
3517 |
op_container_end_page |
3530 |
_version_ |
1766216571253948416 |