DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS
The accuracy of digital elevation models (DEMs) in urban areas is influenced by numerous factors including land cover and terrain irregularities. Moreover, building artefacts in global DEMs cause artificial blocking of surface flow pathways. This compromises their quality and adequacy for hydrologic...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2024-03-01
|
Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/275/2024/isprs-archives-XLVIII-4-W9-2024-275-2024.pdf |
_version_ | 1797268062978703360 |
---|---|
author | C. Okolie C. Okolie C. Okolie J. Mills A. Adeleke J. Smit |
author_facet | C. Okolie C. Okolie C. Okolie J. Mills A. Adeleke J. Smit |
author_sort | C. Okolie |
collection | DOAJ |
description | The accuracy of digital elevation models (DEMs) in urban areas is influenced by numerous factors including land cover and terrain irregularities. Moreover, building artefacts in global DEMs cause artificial blocking of surface flow pathways. This compromises their quality and adequacy for hydrological and environmental modelling in urban landscapes where precise and accurate terrain information is needed. In this study, the extreme gradient boosting (XGBoost) ensemble algorithm is adopted for enhancing the accuracy of two medium-resolution 30-metre DEMs over Cape Town, South Africa: Copernicus GLO-30 and ALOS World 3D (AW3D). XGBoost is a scalable, portable and versatile gradient boosting library that can solve many environmental modelling problems. The training datasets are comprised of eleven predictor variables including elevation, urban footprints, slope, aspect, surface roughness, topographic position index, terrain ruggedness index, terrain surface texture, vector roughness measure, forest cover and bare ground cover. The target variable (elevation error) was calculated with respect to highly accurate airborne LiDAR. After training and testing, the model was applied for correcting the DEMs at two implementation sites. The corrections achieved significant accuracy gains which are competitive with other proposed methods. There was a 46–53% reduction in the root mean square error (RMSE) of Copernicus DEM, and a 72–73% reduction in the RMSE of AW3D DEM. These results showcase the potential of gradient-boosted decision trees for enhancing the quality of global DEMs, especially in urban areas. |
first_indexed | 2024-04-25T01:26:31Z |
format | Article |
id | doaj.art-22f570cce7b1491daf494525419204c5 |
institution | Directory Open Access Journal |
issn | 1682-1750 2194-9034 |
language | English |
last_indexed | 2024-04-25T01:26:31Z |
publishDate | 2024-03-01 |
publisher | Copernicus Publications |
record_format | Article |
series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
spelling | doaj.art-22f570cce7b1491daf494525419204c52024-03-08T19:33:11ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342024-03-01XLVIII-4-W9-202427528210.5194/isprs-archives-XLVIII-4-W9-2024-275-2024DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERSC. Okolie0C. Okolie1C. Okolie2J. Mills3A. Adeleke4J. Smit5Division of Geomatics, University of Cape Town, South AfricaDepartment of Surveying & Geoinformatics, University of Lagos, NigeriaSchool of Engineering, Newcastle University, United KingdomSchool of Engineering, Newcastle University, United KingdomDepartment of Geography, Geoinformatics and Meteorology, University of Pretoria, South AfricaCape Peninsula University of Technology, South AfricaThe accuracy of digital elevation models (DEMs) in urban areas is influenced by numerous factors including land cover and terrain irregularities. Moreover, building artefacts in global DEMs cause artificial blocking of surface flow pathways. This compromises their quality and adequacy for hydrological and environmental modelling in urban landscapes where precise and accurate terrain information is needed. In this study, the extreme gradient boosting (XGBoost) ensemble algorithm is adopted for enhancing the accuracy of two medium-resolution 30-metre DEMs over Cape Town, South Africa: Copernicus GLO-30 and ALOS World 3D (AW3D). XGBoost is a scalable, portable and versatile gradient boosting library that can solve many environmental modelling problems. The training datasets are comprised of eleven predictor variables including elevation, urban footprints, slope, aspect, surface roughness, topographic position index, terrain ruggedness index, terrain surface texture, vector roughness measure, forest cover and bare ground cover. The target variable (elevation error) was calculated with respect to highly accurate airborne LiDAR. After training and testing, the model was applied for correcting the DEMs at two implementation sites. The corrections achieved significant accuracy gains which are competitive with other proposed methods. There was a 46–53% reduction in the root mean square error (RMSE) of Copernicus DEM, and a 72–73% reduction in the RMSE of AW3D DEM. These results showcase the potential of gradient-boosted decision trees for enhancing the quality of global DEMs, especially in urban areas.https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/275/2024/isprs-archives-XLVIII-4-W9-2024-275-2024.pdf |
spellingShingle | C. Okolie C. Okolie C. Okolie J. Mills A. Adeleke J. Smit DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
title | DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS |
title_full | DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS |
title_fullStr | DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS |
title_full_unstemmed | DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS |
title_short | DIGITAL ELEVATION MODEL CORRECTION IN URBAN AREAS USING EXTREME GRADIENT BOOSTING, LAND COVER AND TERRAIN PARAMETERS |
title_sort | digital elevation model correction in urban areas using extreme gradient boosting land cover and terrain parameters |
url | https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/275/2024/isprs-archives-XLVIII-4-W9-2024-275-2024.pdf |
work_keys_str_mv | AT cokolie digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters AT cokolie digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters AT cokolie digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters AT jmills digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters AT aadeleke digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters AT jsmit digitalelevationmodelcorrectioninurbanareasusingextremegradientboostinglandcoverandterrainparameters |