A Mamba-Based Multimodal Network for Multiscale Blast-Induced Rapid Structural Damage Assessment

Wanli Ma; Sivasakthy Selvakumaran; Dain G. Farrimond; Adam A. Dennis; Samuel E. Rigby

cs.AIApr 13, 2026

A Mamba-Based Multimodal Network for Multiscale Blast-Induced Rapid Structural Damage Assessment

Authors: Wanli Ma, Sivasakthy Selvakumaran, Dain G. Farrimond, Adam A. Dennis, Samuel E. Rigby

Organizations: Department of Engineering, University of Cambridge, Cambridge CB2 1TN, UK · School of Mechanical, Aerospace and Civil Engineering, University of Sheffield, Sheffield S10 2TN, UK

Abstract

Accurate and rapid structural damage assessment (SDA) is crucial for post-disaster management, helping responders prioritise resources, plan rescues, and support recovery. Traditional field inspections, though precise, are limited by accessibility, safety risks, and time constraints, especially after large explosions. Machine learning with remote sensing has emerged as a scalable solution for rapid SDA, with Mamba-based networks achieving state-of-the-art performance. However, these methods often require extensive training and large datasets, limiting real-world applicability. Moreover, they fail to incorporate key physical characteristics of blast loading for SDA. To overcome these challenges, we propose a Mamba-based multimodal network for rapid SDA that integrates multi-scale blast-loading information with optical remote sensing images. Evaluated on the 2020 Beirut explosion, our method significantly improves performance over state-of-the-art approaches. Code is available at: https://github.com/IMPACTSquad/Blast-Mamba

Figures & tables

Fig. 1: The workflow of the proposed rapid structural damage assessment method. BS encoder denotes the building segmentation decoder, and SDA encoder denotes the structural damage assessment decoder.

Fig. 2: Structure of the fine-tuning framework. BS: building segmentation; RSI: remote sensing images.

Fig. 3: Residual attention–base spatiotemporal state space (RA-STSS) module

Type	Method	$F_{1}^{\text{loc}}$	$F_{1}^{\text{clf}}$	$F_{1}^{\text{overall}}$	Damage $F_{1}$ per class
					Intact	Damaged	Destroyed
$\mathcal{C}$	DeepLabV3+ [ 9 ]	81.43	70.61	73.85	77.66	53.78	90.76
	UNet [ 24 ]	82.32	51.79	60.95	73.82	30.75	84.28
	SiamAttnUNet [ 1 ]	82.05	60.56	67.01	71.79	41.79	85.63
	SiamCRNN [ 8 ]	85.66	71.74	75.91	84.90	51.03	95.73
$\mathcal{T}$	DamFormer [ 5 ]	85.14	79.54	81.22	87.03	63.18	96.16

TABLE I: Performance comparison between the proposed method and state-of-the-art methods on the Blast-7 dataset. The symbols

\mathcal{C}

\mathcal{T}

, and

\mathcal{M}

represent convolutional neural network (CNN)–based methods, Transformer-based methods, and Mamba-based methods, respectively.

Method	$F_{1}^{loc}$	$F_{1}^{clf}$	$F_{1}^{overall}$	Damage $F_{1}$ per class
				Intact	Damaged	Destroyed
Pretrain	78.50	1.39	24.52	81.09	0.47	27.70
FT	88.70	84.81	85.98	91.21	71.96	95.11
FT w/ D	89.09	84.96	86.20	91.55	71.80	95.58
FT w/ B	88.98	88.30	88.50	93.54	77.96	95.64

TABLE II: Ablation study on fine-tuning and the use of distance and blast information for SDA on the Blast-7 dataset. FT: fine-tuning; w/D: with distance; w/B: with blast loading.

Fig. 4: Comparison of training time versus performance ( $F^{loc}{1}$ , $F^{clf}{1}$ , $F^{overall}_{1}$ ) of the proposed and existing methods for Blast-7 structural damage assessment.

Disaster Damage Assessment

Disaster damage assessment uses imagery and geospatial data to locate, classify, and quantify damage for emergency response and recovery.

See all 49 papers

Multi-Modal Attention for Automated Disaster Damage Assessment Using Remote Sensing Imagery and Deep Learning
Jun 12, 2026Tewodros Syum Gebre, Jagrati Talreja, Leila Hashemi-BeniDisaster Damage Assessment Remote Sensing Change Detection
Built Environment Department, College of Science and Technology, North Carolina A&T State University, Greensboro, NC, USA · United Nations University Institute for Water, Environment and Health, Richmond Hill, ON, Canada
Improved MambdaBDA Framework for Robust Building Damage Assessment Across Disaster Domains
Mar 1, 2026Alp Eren Gençoğlu, Hazım Kemal EkenelDisaster Damage Assessment Building Damage Assessment
Department of Computer Engineering, Istanbul Technical University, Türkiye · Division of Engineering, New York University Abu Dhabi, UAE
RAPID: A Reproducible Multi-Agent Pipeline for Interpretable Disaster Damage Assessment from Satellite and Street-View Imagery
Jun 20, 2026Yifan Yang, Wenjing Gong, Kaili Zhang +5Remote Sensing Image Understanding Disaster Damage Assessment
Texas A&M University College Station, Texas, USA · National University of Singapore Singapore, Singapore · The University of Alabama Tuscaloosa, Alabama, USA

Jun 12, 2026J/K move · Enter open · S save

Multi-Modal Attention for Automated Disaster Damage Assessment Using Remote Sensing Imagery and Deep Learning

Tewodros Syum Gebre, Jagrati Talreja, Leila Hashemi-Beni

Built Environment Department, College of Science and Technology, North Carolina A&T State University, Greensboro, NC, USA · United Nations University Institute for Water, Environment and Health, Richmond Hill, ON, Canada

Disaster Damage Assessment Remote Sensing Change Detection Multimodal Classification

Open paper AskPDF arXiv

Timely and accurate disaster damage assessment is crucial for effective emergency response, resource allocation, and recovery. Traditional methods, which often rely on manual inspections or sparse data, are typically slow and error-prone. This paper introduces a novel framework leveraging remote sensing imagery and deep learning to automate building damage classification. Using pre- and post-disaster satellite imagery, our model categorizes buildings into four damage levels: no damage, minor damage, major damage, and destroyed. The core innovation is a multi-modal attention mechanism that fuses bi-temporal features to explicitly detect and assess structural changes. We employ a lightweight ConvNeXT-Tiny backbone to ensure efficient processing without compromising performance. Key contributions include: (1) a cross-attention module for multi-modal data fusion, (2) an optimized preprocessing pipeline for large-scale datasets, and (3) robust data augmentation techniques. Experiments on a large-scale disaster dataset demonstrate an overall classification accuracy of 94.90%. The model effectively discriminates between damage categories and remains resilient to incomplete data. This system significantly improves assessment speed and accuracy, aiding emergency responders in prioritizing interventions. This work advances automated disaster damage detection by integrating multi-temporal imagery with deep learning, offering a scalable solution for real-time response.

Abstract

Disaster Damage Assessment

Explore similar work

Multi-Modal Attention for Automated Disaster Damage Assessment Using Remote Sensing Imagery and Deep Learning

Improved MambdaBDA Framework for Robust Building Damage Assessment Across Disaster Domains

RAPID: A Reproducible Multi-Agent Pipeline for Interpretable Disaster Damage Assessment from Satellite and Street-View Imagery