Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

ViTConvKAN : a hybrid transformer–KAN modeling for post-earthquake building damage mapping using VHR imagery

Khankeshizadeh, Seyed Ehsan LU ; Mohammadzadeh, Ali and Jamali, Sadegh LU orcid (2026) In Bulletin of Earthquake Engineering
Abstract

Accurate mapping of earthquake-damaged buildings is essential for rapid post-disaster assessment and urban recovery planning. Very high-resolution (VHR) satellite imagery enables large-scale geospatial damage detection; however, conventional convolutional neural networks (CNNs) often struggle to model complex spatial dependencies and nonlinear damage patterns efficiently. To address these limitations, this study proposes ViTConvKAN, a hybrid architecture that combines vision transformers (ViTs) for long-range spatial modeling with convolutional Kolmogorov–Arnold networks (ConvKANs) for nonlinear feature representation. The framework includes data preprocessing, supervised training, building-level damage classification, and quantitative... (More)

Accurate mapping of earthquake-damaged buildings is essential for rapid post-disaster assessment and urban recovery planning. Very high-resolution (VHR) satellite imagery enables large-scale geospatial damage detection; however, conventional convolutional neural networks (CNNs) often struggle to model complex spatial dependencies and nonlinear damage patterns efficiently. To address these limitations, this study proposes ViTConvKAN, a hybrid architecture that combines vision transformers (ViTs) for long-range spatial modeling with convolutional Kolmogorov–Arnold networks (ConvKANs) for nonlinear feature representation. The framework includes data preprocessing, supervised training, building-level damage classification, and quantitative evaluation. The method is validated on post-event VHR datasets from Haiti-Port-au-Prince and Iran-Bam, including cross-region transfer experiments under zero-shot and few-shot fine-tuning settings. Across both regions, ViTConvKAN achieves an average OA of 88.09% and a KC of 73.07%, outperforming several established CNN-based baselines. The results indicate that integrating transformer-based contextual modeling with nonlinear ConvKAN feature extraction improves building-level damage discrimination while maintaining scalable deployment characteristics. This hybrid approach provides a robust framework for post-earthquake damage mapping using VHR imagery.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
epub
subject
keywords
Deep learning, Earthquake-damaged buildings, KAN, Post-earthquake damage assessment, Remote sensing, Very high-resolution satellite imagery, Vision transformer
in
Bulletin of Earthquake Engineering
publisher
Springer Science and Business Media B.V.
external identifiers
  • scopus:105041072969
ISSN
1570-761X
DOI
10.1007/s10518-026-02530-9
language
English
LU publication?
yes
id
276a7805-6fc1-46a6-a139-6594edf389e1
date added to LUP
2026-07-01 11:57:11
date last changed
2026-07-01 15:12:10
@article{276a7805-6fc1-46a6-a139-6594edf389e1,
  abstract     = {{<p>Accurate mapping of earthquake-damaged buildings is essential for rapid post-disaster assessment and urban recovery planning. Very high-resolution (VHR) satellite imagery enables large-scale geospatial damage detection; however, conventional convolutional neural networks (CNNs) often struggle to model complex spatial dependencies and nonlinear damage patterns efficiently. To address these limitations, this study proposes ViTConvKAN, a hybrid architecture that combines vision transformers (ViTs) for long-range spatial modeling with convolutional Kolmogorov–Arnold networks (ConvKANs) for nonlinear feature representation. The framework includes data preprocessing, supervised training, building-level damage classification, and quantitative evaluation. The method is validated on post-event VHR datasets from Haiti-Port-au-Prince and Iran-Bam, including cross-region transfer experiments under zero-shot and few-shot fine-tuning settings. Across both regions, ViTConvKAN achieves an average OA of 88.09% and a KC of 73.07%, outperforming several established CNN-based baselines. The results indicate that integrating transformer-based contextual modeling with nonlinear ConvKAN feature extraction improves building-level damage discrimination while maintaining scalable deployment characteristics. This hybrid approach provides a robust framework for post-earthquake damage mapping using VHR imagery.</p>}},
  author       = {{Khankeshizadeh, Seyed Ehsan and Mohammadzadeh, Ali and Jamali, Sadegh}},
  issn         = {{1570-761X}},
  keywords     = {{Deep learning; Earthquake-damaged buildings; KAN; Post-earthquake damage assessment; Remote sensing; Very high-resolution satellite imagery; Vision transformer}},
  language     = {{eng}},
  publisher    = {{Springer Science and Business Media B.V.}},
  series       = {{Bulletin of Earthquake Engineering}},
  title        = {{ViTConvKAN : a hybrid transformer–KAN modeling for post-earthquake building damage mapping using VHR imagery}},
  url          = {{http://dx.doi.org/10.1007/s10518-026-02530-9}},
  doi          = {{10.1007/s10518-026-02530-9}},
  year         = {{2026}},
}