EUVIMEDEUVIMED entdecken
Crossref

Hyperspectral image classification using a large selective kernel network hybridized tokenization transformer

Xiaoqing Wan, Yupeng He, Weizhe Gao, Feng Chen, Wenhui Chen

Journal of the Optical Society of America B · 01.02.2025 · 42 (2) · 251

DOI: 10.1364/josab.537962 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Optical Society of America B, online erschienen am 13.01.2025. Verfasst von Xiaoqing Wan und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 35 Literaturangaben. Gefördert von: Hengyang Normal University, Hengyang City Guidance Plan Project, 14th Five-Year Plan Key Disciplines and Application-oriented Special Disciplines of Hunan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Convolutional neural networks (CNNs) are widely used for hyperspectral image (HSI) classification. However, the high spatial and spectral dimensionality of HSIs often leads to significant computational costs and challenges during network training. Moreover, CNNs are limited in capturing high-level semantic features. In contrast, transformer models are better suited to modeling high-level semantic information and capturing long-range dependencies, making them a promising approach for HSI classification. In this paper, we propose a novel HSI classification framework, LSKTT, which integrates a large selective kernel (LSK) network with a tokenization transformer. The framework consists of three key components: (1) dimensionality reduction via principal component analysis (PCA) to reduce spectral bands and feature extraction through a hybrid 3D and 2D convolutional network; (2) contextual modeling using the LSK network, which dynamically adjusts its receptive field to capture spatial information at varying scales; and (3) a feature tokenization transformer that learns abstract spatial–spectral features to capture deeper semantic relationships. Experimental results on three real-world hyperspectral image datasets demonstrate that the proposed LSKTT outperforms several state-of-the-art methods, both qualitatively and quantitatively.

Förderung

  • Hengyang Normal University
  • Hengyang City Guidance Plan Project
  • 14th Five-Year Plan Key Disciplines and Application-oriented Special Disciplines of Hunan Province
  • Science and Technology Plan Project of Hunan Province
  • Scientific Research Fund of Hunan Provincial Education Department

Zitieren

APA

Wan, X., He, Y., Gao, W., Chen, F., & Chen, W. (2025). Hyperspectral image classification using a large selective kernel network hybridized tokenization transformer. Journal of the Optical Society of America B, 42(2), 251. https://doi.org/10.1364/josab.537962

Vancouver

Wan X, He Y, Gao W, Chen F, Chen W. Hyperspectral image classification using a large selective kernel network hybridized tokenization transformer. Journal of the Optical Society of America B. 2025;42(2):251. doi:10.1364/josab.537962

BibTeX
@article{Wan2025,
  author = {Wan, Xiaoqing and He, Yupeng and Gao, Weizhe and Chen, Feng and Chen, Wenhui},
  title = {Hyperspectral image classification using a large selective kernel network hybridized tokenization transformer},
  journal = {Journal of the Optical Society of America B},
  year = {2025},
  volume = {42},
  number = {2},
  pages = {251},
  doi = {10.1364/josab.537962}
}
Kostenlos starten →