DeepC4: Deep Conditional Census-Constrained Clustering for Large-scale Multitask Spatial Disaggregation of Urban Morphology

2025-07-30

AI Summary

To address low spatial discretization accuracy of urban morphology under weak supervision in developing countries, inconsistencies between local mapping and census statistics, and severe model uncertainty propagation, this paper proposes the Deep Conditional Census-Constrained Clustering (DC³) framework. DC³ embeds coarse-grained census statistics as hard cluster-level constraints into a multi-task deep learning pipeline, jointly modeling high-resolution remote sensing imagery and derived features—thereby preserving pixel-level discriminability while enforcing regional statistical consistency. Its key innovation lies in the first integration of conditional label relationship modeling with census-driven constraints within a clustering architecture, enhancing both local coherence and physical interpretability. Evaluated across Rwanda’s three-tier administrative units, DC³ significantly outperforms GEM and METEOR, reducing errors in building exposure and physical vulnerability mapping by 23.6%, thereby effectively supporting spatial auditing for SDGs 11 and 13.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Clustering

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

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.

Problem

• Addresses local discrepancy in urban morphology mapping
• Improves accuracy using census statistics as constraints
• Enhances large-scale spatial disaggregation with deep learning

Innovation

• Deep learning for spatial disaggregation
• Census statistics as cluster constraints
• Multitask learning with conditional labels