GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation
Remote Sensing · 24.07.2026 · 18 (15) · 2449
DOI: 10.3390/rs18152449 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 24.07.2026. Verfasst von Sangning Li und 5 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Key Laboratory of Land Satellite Remote Sensing Application, Ministry of Natural Resources of the People’s Republic of China.
Abstract (Originalsprache)
Accurate pavement crack segmentation is critical for road infrastructure assessment, yet it remains challenging due to complex background noise and highly variable crack topologies. While emerging visual Mamba models excel in long-range contextual modeling, their inherent 1D sequence flattening process compromises local 2D spatial continuity. To address this limitation, we propose GeoMamba, a geometric-prior-infused multi-scale deformable visual Mamba network for road crack semantic segmentation. First, we design a Multi-Scale Deformable Visual State Space (MDVSS) module to extract multi-scale contextual features and dynamically adapt to tortuous crack paths through a novel deformable scanning mechanism. Second, a Geometric-Topology Prior Injection (GTPI) module is introduced to mitigate serialization artifacts. By leveraging deterministic, parameter-free analytical operators (i.e., Sobel and Laplace), the GTPI module explicitly extracts and adaptively infuses multi-scale structural priors into the Mamba decoder via gated skip connections, intrinsically reconstructing crack typologies while suppressing pseudo-structural noise. Comprehensive experiments on DeepCrack and Concrete3K datasets demonstrate that GeoMamba outperforms nine state-of-the-art methods. Specifically, it achieves peak performance on the DeepCrack dataset with an mIoU of 83.79% and an F1 score of 89.27%, demonstrating exceptional semantic segmentation performance, superior topological continuity, and robust generalization across diverse pavement materials.
Förderung
- National Natural Science Foundation of China
- Key Laboratory of Land Satellite Remote Sensing Application, Ministry of Natural Resources of the People’s Republic of China
Zitieren
APA
Li, S., Liu, B., Guan, H., Ma, L., Yu, Y., & Xu, Y. (2026). GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation. Remote Sensing, 18(15), 2449. https://doi.org/10.3390/rs18152449
Vancouver
Li S, Liu B, Guan H, Ma L, Yu Y, Xu Y. GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation. Remote Sensing. 2026;18(15):2449. doi:10.3390/rs18152449
BibTeX
@article{Li2026,
author = {Li, Sangning and Liu, Bin and Guan, Haiyan and Ma, Lingfei and Yu, Yongtao and Xu, Yongming},
title = {GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {15},
pages = {2449},
doi = {10.3390/rs18152449}
}
EUVIMED