Building height and roof material dataset - Dominica, Pichelin and Roseau, 2018, 2024

<table style="border-collapse: collapse; margin-left: 4.8pt; margin-right: 4.8pt; width: 99%;"> <tbody> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Science Case Name</span></strong></p> </td> <td style="padding: 5pt;"> <p>&nbsp;Multi-hazards in the Caribbean SIDS</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Dataset Title</span></strong></p> </td> <td style="padding: 5pt;"> <p>Building height and roof material dataset - Dominica, Pichelin and Roseau, 2018, 2024</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Dataset Description</span></strong></p> </td> <td style="padding: 5pt;"> <p>Building height predictions of a monocular depth estimation network trained on UAV imageries of Pichelin (2024) and Roseau (2018) trained on Google&rsquo;s Open Building 2.5D Temporal dataset ( https://sites.research.google/gr/open-buildings/temporal/) and roof material classification using the model provided in GFDRR&rsquo;s Caribbean Rooftop Classification (https://github.com/GFDRR/caribbean-rooftop-classification):</p> <p>&middot;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; pichelin_network_height_rooftype_predictions.gpkg <a name="OLE_LINK3"></a>&ndash; Geopackage containing building height estimation and roof material classification of Pichelin, Dominica (2024).</p> <p>&middot;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; roseau_network_predicitons.gpkg &ndash; Geopackage containing building height estimation and roof material classification of Roseau, Dominica (2018).</p> <p>&nbsp;</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Key Methodologies</span></strong></p> </td> <td style="padding: 5pt;"> <p>A monocular depth estimation network called MiDaS (https://pytorch.org/hub/intelisl_midas_v2/) and the roof type and material classification network of GFDRR based on ResNet-50.</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Temporal Domain</span></strong></p> </td> <td style="padding: 5pt;"> <p><span lang="NL">2018 and 2024</span></p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Spatial Domain</span></strong></p> </td> <td style="padding: 5pt;"> <p>Roseau and Pichelin, Dominica</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Key Indicators</span></strong></p> </td> <td style="padding: 5pt;"> <p>Building height and roof material</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Data Format</span></strong></p> </td> <td style="padding: 5pt;"> <p>A vector as a Geopackage (.gpkg)</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Source Data</span></strong></p> </td> <td style="padding: 5pt;"> <p>Openaerialmap UAV image of Roseau (2018), UAV images of Pichelin (2024) captured using a DJI drone, and Google&rsquo;s Open Building 2.5D Temporal dataset</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Accessibility</span></strong></p> </td> <td style="padding: 5pt;"> <p><strong>10.5281/zenodo.17100331</strong></p> <p><strong>Disclaimer</strong><em> &ndash; This dataset is provided solely for research purposes. It is based on available data and has not been verified or approved by any local authority. The dataset represents the outcome of the applied methodology and research judgments, and does not reflect the official policies or positions of the authorities of Dominica. </em></p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Stakeholder Relevance</span></strong></p> </td> <td style="padding: 5pt;"> <p>The building height and roof material dataset can be used for hazard mapping and risk management. For example, they can be integrated into the digital twin environment of buildings-at-risk.</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Limitations/Assumptions</span></strong></p> </td> <td style="padding: 5pt;"> <p>Temporal alignment of images used and building height dataset the network was trained on.</p> </td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Additional Information</span></strong></p> </td> <td style="padding: 5pt;">NA</td> </tr> <tr> <td style="padding: 5pt;"> <p><strong><span lang="NL">Contact Information</span></strong></p> </td> <td style="padding: 5pt;"> <p>Bergado, John Ray (UT-ITC)</p> </td> </tr> </tbody> </table> <p>&nbsp;</p>