DeepC4: Deep Conditional Census-Constrained Clustering for Large-scale Multitask Spatial Disaggregation of Urban Morphology
This work addresses the challenge of generating accurate, fine-grained urban morphology maps for sustainable development and disaster risk reduction in developing regions, offering a novel deep learning technique to enhance spatial auditing of coarse data.
The paper tackled the problem of large-scale spatial disaggregation of urban morphology in developing economies, where classical methods face local discrepancies and uncertainties due to weak supervision; it introduced DeepC4, a deep learning approach that improved the quality of Rwandan maps for building exposure and vulnerability at the third-level administrative unit compared to existing methods like GEM and METEOR.
To understand our global progress for sustainable development and disaster risk reduction in many developing economies, two recent major initiatives - the Uniform African Exposure Dataset of the Global Earthquake Model (GEM) Foundation and the Modelling Exposure through Earth Observation Routines (METEOR) Project - implemented classical spatial disaggregation techniques to generate large-scale mapping of urban morphology using the information from various satellite imagery and its derivatives, geospatial datasets of the built environment, and subnational census statistics. However, the local discrepancy with well-validated census statistics and the propagated model uncertainties remain a challenge in such coarse-to-fine-grained mapping problems, specifically constrained by weak and conditional label supervision. Therefore, we present Deep Conditional Census-Constrained Clustering (DeepC4), a novel deep learning-based spatial disaggregation approach that incorporates local census statistics as cluster-level constraints while considering multiple conditional label relationships in a joint multitask learning of the patterns of satellite imagery. To demonstrate, compared to GEM and METEOR, we enhanced the quality of Rwandan maps of urban morphology, specifically building exposure and physical vulnerability, at the third-level administrative unit from the 2022 census. As the world approaches the conclusion of our global frameworks in 2030, our work has offered a new deep learning-based mapping technique towards a spatial auditing of our existing coarse-grained derived information at large scales.