EUVIMEDEUVIMED entdecken
Crossref

SegHSI: Semantic Segmentation of Hyperspectral Images With Limited Labeled Pixels

Huan Liu, Wei Li, Xiang-Gen Xia, Mengmeng Zhang, Zhengqi Guo, Lujie Song

IEEE Transactions on Image Processing · 2024 · 33 · 6469-6482

DOI: 10.1109/tip.2024.3492724 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2024. Verfasst von Huan Liu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 40-mal zitiert (Crossref) und nennt 66 Literaturangaben. Gefördert von: National Key Laboratory of Multispectral Information Intelligent Processing Technology, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral images (HSIs), with hundreds of narrow spectral bands, are increasingly used for ground object classification in remote sensing. However, many HSI classification models operate pixel-by-pixel, limiting the utilization of spatial information and resulting in increased inference time for the whole image. This paper proposes SegHSI, an effective and efficient end-to-end HSI segmentation model, alongside a novel training strategy. SegHSI adopts a head-free structure with cluster attention modules and spatial-aware feedforward networks (SA-FFN) for multiscale spatial encoding. Cluster attention encodes pixels through constructed clusters within the HSI, while SA-FFN integrates depth-wise convolution to enhance spatial context. Our training strategy utilizes a student-teacher model framework that combines labeled pixel class information with consistency learning on unlabeled pixels. Experiments on three public HSI datasets demonstrate that SegHSI not only surpasses other state-of-the-art models in segmentation accuracy but also achieves inference time at the scale of seconds, even reaching sub-second speeds for full-image classification. Code is available at https://github.com/huanliu233/SegHSI.

Förderung

  • National Key Laboratory of Multispectral Information Intelligent Processing Technology
  • National Natural Science Foundation of China

Zitieren

APA

Liu, H., Li, W., Xia, X. G., Zhang, M., Guo, Z., & Song, L. (2024). SegHSI: Semantic Segmentation of Hyperspectral Images With Limited Labeled Pixels. IEEE Transactions on Image Processing, 33, 6469-6482. https://doi.org/10.1109/tip.2024.3492724

Vancouver

Liu H, Li W, Xia XG, Zhang M, Guo Z, Song L. SegHSI: Semantic Segmentation of Hyperspectral Images With Limited Labeled Pixels. IEEE Transactions on Image Processing. 2024;33:6469-6482. doi:10.1109/tip.2024.3492724

BibTeX
@article{Liu2024,
  author = {Liu, Huan and Li, Wei and Xia, Xiang-Gen and Zhang, Mengmeng and Guo, Zhengqi and Song, Lujie},
  title = {SegHSI: Semantic Segmentation of Hyperspectral Images With Limited Labeled Pixels},
  journal = {IEEE Transactions on Image Processing},
  year = {2024},
  volume = {33},
  pages = {6469--6482},
  doi = {10.1109/tip.2024.3492724}
}
Kostenlos starten →