EUVIMEDEUVIMED entdecken
Crossref

S2CMamba: Structure-to-Contrast Semantic Mamba for Hyperspectral and LiDAR Joint Classification

Niannian Zhang, Thomas Xinzhang Wu, Lianhui Liang, Bing Tu, Jun Li, Antonio Plaza

IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 2002916-2002916

DOI: 10.1109/tgrs.2026.3696285 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Niannian Zhang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Guangxi Science and Technology Major Program, Guangxi Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral images (HSI) and light detection and ranging (LiDAR) provide complementary spectral and elevation information, making their fusion an effective solution for multimodal remote sensing (RS) classification. However, existing fusion methods often fail to adequately capture complex spatial topology and long-range contextual dependencies. Moreover, foreground–background semantic entanglement and the lack of semantic-guided fusion mechanisms further degrade feature discrimination. To address these challenges, a novel structure-to-contrast semantic Mamba (S2CMamba) network is proposed for the joint classification of HSI and LiDAR data. Specifically, a structure-aware state space module (S2SSM) module is designed for both HSI and LiDAR branches to encode multidirectional spatial structures and capture long-range global spatial dependencies within each modality. Then, the proposed semantic decoupling unit (SDU) explicitly separates foreground and background semantic responses in the fused HSI–LiDAR representations, which effectively suppresses background interference and alleviates semantic ambiguity during multimodal feature fusion. Based on the decoupled semantic priors, a contrast-guided feature aggregation module (CGAM) is developed to enable contrast-modulated local window aggregation over the fused HSI–LiDAR features, thereby improving intra-class compactness and enhancing class-boundary discrimination. Through experiments conducted on three commonly used HSI and LiDAR datasets, the performance of S2CMamba was consistently demonstrated to surpass that of state-of-the-art (SOTA) methods for HSI and LiDAR.

Förderung

  • National Natural Science Foundation of China
  • Guangxi Science and Technology Major Program
  • Guangxi Natural Science Foundation
  • Opening Fund of the Key Laboratory for Coastal Marine Eco-Environment Processes and Carbon Sinks of Hainan Province
  • Project for Enhancing Young and Middle-Aged Teacher’s Research Basis Ability in Colleges of Guangxi

Zitieren

APA

Zhang, N., Wu, T. X., Liang, L., Tu, B., Li, J., & Plaza, A. (2026). S2CMamba: Structure-to-Contrast Semantic Mamba for Hyperspectral and LiDAR Joint Classification. IEEE Transactions on Geoscience and Remote Sensing, 64, 2002916-2002916. https://doi.org/10.1109/tgrs.2026.3696285

Vancouver

Zhang N, Wu TX, Liang L, Tu B, Li J, Plaza A. S2CMamba: Structure-to-Contrast Semantic Mamba for Hyperspectral and LiDAR Joint Classification. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:2002916-2002916. doi:10.1109/tgrs.2026.3696285

BibTeX
@article{Zhang2026,
  author = {Zhang, Niannian and Wu, Thomas Xinzhang and Liang, Lianhui and Tu, Bing and Li, Jun and Plaza, Antonio},
  title = {S2CMamba: Structure-to-Contrast Semantic Mamba for Hyperspectral and LiDAR Joint Classification},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  volume = {64},
  pages = {2002916--2002916},
  doi = {10.1109/tgrs.2026.3696285}
}
Kostenlos starten →