ChangeQuery: Advancing Remote Sensing Change Analysis for Natural and Human-Induced Disasters from Visual Detection to Semantic Understanding

2026-04-24

AI Summary

This work addresses the limitations of existing remote sensing–based disaster analysis methods, which are often confined to single-modality optical data, natural disaster scenarios, and lack interactive semantic understanding for complex strategic queries. To overcome these challenges, the authors propose a unified multimodal framework that integrates pre-disaster optical semantic information with post-disaster SAR structural features, enabling query-driven precise damage quantification, regional description, and situational summarization. Key contributions include the construction of DICQ—the first dataset encompassing both natural disasters and human-induced conflicts—alongside a novel “statistics-first, generation-after” automated semantic annotation pipeline. The framework further introduces multimodal fusion, vision–language interactive reasoning, and hierarchical instruction generation mechanisms, achieving state-of-the-art performance in complex disaster monitoring while delivering robust, interpretable, all-weather, and multi-hazard remote sensing situational awareness.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Rapid situational awareness is critical in post-disaster response. While remote sensing damage assessment is evolving from pixel-level change detection to high-level semantic analysis, existing vision-language methodologies still struggle to provide actionable intelligence for complex strategic queries. They remain severely constrained by unimodal optical dependence, a prevailing bias towards natural disasters, and a fundamental lack of grounded interactivity. To address these limitations, we present ChangeQuery, a unified multimodal framework designed for comprehensive, all-weather disaster situation awareness. To overcome modality constraints and scenario biases, we construct the Disaster-Induced Change Query (DICQ) dataset, a large-scale benchmark coupling pre-event optical semantics with post-event SAR structural features across a balanced distribution of natural catastrophes and armed conflicts. Furthermore, to provide the high-quality supervision required for interactive reasoning, we propose a novel Automated Semantic Annotation Pipeline. Adhering to a ``statistics-first, generation-later'' paradigm, this engine automatically transforms raw segmentation masks into grounded, hierarchical instruction sets, effectively equipping the model with fine-grained spatial and quantitative awareness. Trained on this structured data, the ChangeQuery architecture operates as an interactive disaster analyst. It supports multi-task reasoning driven by diverse user queries, delivering precise damage quantification, region-specific descriptions, and holistic post-disaster summaries. Extensive experiments demonstrate that ChangeQuery establishes a new state-of-the-art, providing a robust and interpretable solution for complex disaster monitoring. The code is available at \href{https://sundongwei.github.io/changequery/}{https://sundongwei.github.io/changequery/}.

Problem

• remote sensing
• change analysis
• disaster response
• multimodal
• semantic understanding

Innovation

• multimodal remote sensing
• semantic change analysis
• interactive reasoning
• SAR-optical fusion
• automated annotation