EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Spectral–Structural Collaborative Learning for Fine-Grained Hyperspectral Mineral Classification

Yichun Qiu iD, Yanshuang Zhang, Shixian Cao, Wenyuan Wu iD, Shanjuan Xie

Sensors · 09.07.2026 · 26 (14) · 4345

DOI: 10.3390/s26144345 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 09.07.2026. Verfasst von Yichun Qiu und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Science and Technology, Hangzhou Normal University, Hangzhou 311121, China. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Zhejiang Provincial Natural Science Foundation, Hangzhou Normal University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fine-grained hyperspectral mineral classification remains challenging due to spectral homogeneity among minerals with different morphologies, severe spectral mixing from intergrowth, and high dimensionality. Existing methods rely on spectral separability assumptions, which become insufficient when spectral differences are subtle and spatial–structural ambiguity is high. To address these limitations, we propose S3AM-ECA-3DCNN, a spectral–structural collaborative feature learning framework. It uses a 3DCNN backbone to jointly model spectral–spatial features with progressive spectral downsampling. The spectral-similarity-based spatial attention module (S3AM) performs spatial purification by suppressing interference from spectrally mixed neighboring regions, and the efficient channel attention (ECA) module adaptively recalibrates discriminative spectral bands to enhance fine-grained representation. This establishes a spatial-first, channel-second collaborative optimization paradigm. To improve generalization under limited training samples, adaptive global pooling and a lightweight classification head are employed to reduce model complexity and mitigate overfitting. Experiments on a 146-class hyperspectral mineral dataset (covering silicates, carbonates, and sulfates) show that the framework achieves 93.424% overall accuracy, 91.099% average accuracy, and a Kappa coefficient (×100) of 93.368, outperforming mainstream methods. It significantly reduces misclassification among spectrally similar but morphologically distinct minerals, demonstrating strong robustness and discriminative capability for large-scale fine-grained tasks.

Schlagwörter

Stichwörter der Autor:innen

collaborative attention paradigmdeep learningfine-grained hyperspectral mineral classificationhyperspectral image classification

Autor:innen und Institutionen

  • School of Information Science and Technology, Hangzhou Normal University, Hangzhou 311121, China.
  • Kharkiv College, Hangzhou Normal University, Hangzhou 311121, China.
  • College of Artificial Intelligence and E-Commerce, Zhejiang Gongshang University Hangzhou College of Commerce, Hangzhou 311121, China.
  • Zhejiang Provincial Key Laboratory of Urban Wetlands and Regional Change, Hangzhou Normal University, Hangzhou 311121, China.

ORCID-Profile: Yichun Qiu, Wenyuan Wu

Förderung

  • National Natural Science Foundation of China
  • Zhejiang Provincial Natural Science Foundation
  • Hangzhou Normal University

Zitieren

APA

Qiu, Y., Zhang, Y., Cao, S., Wu, W., & Xie, S. (2026). Spectral–Structural Collaborative Learning for Fine-Grained Hyperspectral Mineral Classification. Sensors, 26(14), 4345. https://doi.org/10.3390/s26144345

Vancouver

Qiu Y, Zhang Y, Cao S, Wu W, Xie S. Spectral–Structural Collaborative Learning for Fine-Grained Hyperspectral Mineral Classification. Sensors. 2026;26(14):4345. doi:10.3390/s26144345

BibTeX
@article{Qiu2026,
  author = {Qiu, Yichun and Zhang, Yanshuang and Cao, Shixian and Wu, Wenyuan and Xie, Shanjuan},
  title = {Spectral–Structural Collaborative Learning for Fine-Grained Hyperspectral Mineral Classification},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {14},
  pages = {4345},
  doi = {10.3390/s26144345}
}
Kostenlos starten →