Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden
Study region: Sweden, a mostly forested country with many small forest roads obstructing topographical modelling of shallow groundwater and streams. Study focus: Maps have traditionally been constructed from aerial photos, but dense forest canopies often obscure these streams from view. Topographica...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | Journal of Hydrology: Regional Studies |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S221458182400497X |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832591767975755776 |
|---|---|
| author | William Lidberg |
| author_facet | William Lidberg |
| author_sort | William Lidberg |
| collection | DOAJ |
| description | Study region: Sweden, a mostly forested country with many small forest roads obstructing topographical modelling of shallow groundwater and streams. Study focus: Maps have traditionally been constructed from aerial photos, but dense forest canopies often obscure these streams from view. Topographical modelling is a widely adopted method for mapping small streams and has been proposed as a potential solution. However, road embankments can disrupt flow paths, acting like dams in the digital landscape. This study presents a novel method where a unique dataset of 28,512 culverts was used to develop a deep learning method to map road culverts, enabling the correction of digital elevation models and enhancing the accuracy of topographically derived stream networks. New hydrological insights for the region: The deep learning model successfully mapped 87 % of all culverts in the test data, and integrating these predicted culverts into topographical models slightly improved the precision of stream networks extracted from high-resolution digital elevation models in the region. These findings demonstrate the possibility for UNet models to enhance hydrological modelling and stream network mapping but further research should focus on reducing the number of false positive culvert predictions. |
| format | Article |
| id | doaj-art-5767756e41b24bc1bdf98ae463fbc0b1 |
| institution | Kabale University |
| issn | 2214-5818 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Hydrology: Regional Studies |
| spelling | doaj-art-5767756e41b24bc1bdf98ae463fbc0b12025-01-22T05:42:13ZengElsevierJournal of Hydrology: Regional Studies2214-58182025-02-0157102148Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in SwedenWilliam Lidberg0Department of Forest Ecology and Management, Swedish University of Agricultural Sciences, Umeå 901 83, SwedenStudy region: Sweden, a mostly forested country with many small forest roads obstructing topographical modelling of shallow groundwater and streams. Study focus: Maps have traditionally been constructed from aerial photos, but dense forest canopies often obscure these streams from view. Topographical modelling is a widely adopted method for mapping small streams and has been proposed as a potential solution. However, road embankments can disrupt flow paths, acting like dams in the digital landscape. This study presents a novel method where a unique dataset of 28,512 culverts was used to develop a deep learning method to map road culverts, enabling the correction of digital elevation models and enhancing the accuracy of topographically derived stream networks. New hydrological insights for the region: The deep learning model successfully mapped 87 % of all culverts in the test data, and integrating these predicted culverts into topographical models slightly improved the precision of stream networks extracted from high-resolution digital elevation models in the region. These findings demonstrate the possibility for UNet models to enhance hydrological modelling and stream network mapping but further research should focus on reducing the number of false positive culvert predictions.http://www.sciencedirect.com/science/article/pii/S221458182400497XStream networkUNetMachine learningLiDARCulvertALS |
| spellingShingle | William Lidberg Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden Journal of Hydrology: Regional Studies Stream network UNet Machine learning LiDAR Culvert ALS |
| title | Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden |
| title_full | Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden |
| title_fullStr | Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden |
| title_full_unstemmed | Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden |
| title_short | Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden |
| title_sort | deep learning enhanced detection of road culverts in high resolution digital elevation models improving stream network accuracy in sweden |
| topic | Stream network UNet Machine learning LiDAR Culvert ALS |
| url | http://www.sciencedirect.com/science/article/pii/S221458182400497X |
| work_keys_str_mv | AT williamlidberg deeplearningenhanceddetectionofroadculvertsinhighresolutiondigitalelevationmodelsimprovingstreamnetworkaccuracyinsweden |