Multi-class Seismic Building Damage Assessment from InSAR Imagery using Quadratic Variational Causal Bayesian Inference

2025-02-25

AI Summary

Addressing challenges in multi-class seismic building damage assessment from InSAR imagery—including strong environmental noise interference, high computational complexity for fine-grained classification, and poor real-time performance at regional scales—this paper proposes the first causal Bayesian inference framework constrained by a quadratic variational lower bound. The method integrates the USGS surface-failure model with building fragility functions, enabling physics-guided causal modeling to decouple deformation signals from noise. A model-driven pruning mechanism is further introduced to accelerate inference. Evaluated on five major earthquake datasets, the framework achieves AUC scores of 0.93–0.96 across all damage levels—up to 35.7% higher than state-of-the-art methods—while reducing computational overhead by over 40%. Critically, it operates without requiring extensive field-collected ground-truth annotations.

Technology Category

• Reasoning under Uncertainty: Probabilistic Inference
• Machine Learning: Bayesian Learning
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Interferometric Synthetic Aperture Radar (InSAR) technology uses satellite radar to detect surface deformation patterns and monitor earthquake impacts on buildings. While vital for emergency response planning, extracting multi-class building damage classifications from InSAR data faces challenges: overlapping damage signatures with environmental noise, computational complexity in multi-class scenarios, and the need for rapid regional-scale processing. Our novel multi-class variational causal Bayesian inference framework with quadratic variational bounds provides rigorous approximations while ensuring efficiency. By integrating InSAR observations with USGS ground failure models and building fragility functions, our approach separates building damage signals while maintaining computational efficiency through strategic pruning. Evaluation across five major earthquakes (Haiti 2021, Puerto Rico 2020, Zagreb 2020, Italy 2016, Ridgecrest 2019) shows improved damage classification accuracy (AUC: 0.94-0.96), achieving up to 35.7% improvement over existing methods. Our approach maintains high accuracy (AUC>0.93) across all damage categories while reducing computational overhead by over 40% without requiring extensive ground truth data.

Problem

• Multi-class building damage assessment
• Separating damage signals from noise
• Improving computational efficiency and accuracy

Innovation

• Quadratic variational causal Bayesian inference
• Integrates InSAR with USGS models
• Reduces computational overhead significantly