EUVIMEDEUVIMED entdecken
Crossref

ITER: Image-to-Pixel Representation for Weakly Supervised HSI Classification

Jiaqi Yang, Bo Du, Di Wang, Liangpei Zhang

IEEE Transactions on Image Processing · 2024 · 33 · 257-272

DOI: 10.1109/tip.2023.3326699 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2024. Verfasst von Jiaqi Yang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 63-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent years have witnessed the superiority of deep learning-based algorithms in the field of HSI classification. However, a prerequisite for the favorable performance of these methods is a large number of refined pixel-level annotations. Due to atmospheric changes, sensor differences, and complex land cover distribution, pixel-level labeling of high-dimensional hyperspectral image (HSI) is extremely difficult, time-consuming, and laborious. To overcome the above hurdle, an Image-To-pixEl Representation (ITER) approach is proposed in this paper. To the best of our knowledge, this is the first time that image-level annotation is introduced to predict pixel-level classification maps for HSI. The proposed model is along the lines of subject modeling to boundary refinement, corresponding to pseudo-label generation and pixel-level prediction. Concretely, in the pseudo-label generation part, the spectral/spatial activation, spectral-spatial alignment loss, and geographic element enhancement are sequentially designed to locate discriminate regions of each category, optimize multi-domain class activation map (CAM) collaborative training, and refine labels, respectively. For the pixel-level prediction portion, a high frequency-aware self-attention in a high-enhanced transformer is put forward to achieve detailed feature representation. With the two-stage pipeline, ITER explores weakly supervised HSI classification with image-level tags, bridging the gap between image-level annotation and dense prediction. Extensive experiments in three benchmark datasets with state-of-the-art (SOTA) works show the performance of the proposed approach.

Förderung

  • National Key Research and Development Program of China
  • National Natural Science Foundation of China

Zitieren

APA

Yang, J., Du, B., Wang, D., & Zhang, L. (2024). ITER: Image-to-Pixel Representation for Weakly Supervised HSI Classification. IEEE Transactions on Image Processing, 33, 257-272. https://doi.org/10.1109/tip.2023.3326699

Vancouver

Yang J, Du B, Wang D, Zhang L. ITER: Image-to-Pixel Representation for Weakly Supervised HSI Classification. IEEE Transactions on Image Processing. 2024;33:257-272. doi:10.1109/tip.2023.3326699

BibTeX
@article{Yang2024,
  author = {Yang, Jiaqi and Du, Bo and Wang, Di and Zhang, Liangpei},
  title = {ITER: Image-to-Pixel Representation for Weakly Supervised HSI Classification},
  journal = {IEEE Transactions on Image Processing},
  year = {2024},
  volume = {33},
  pages = {257--272},
  doi = {10.1109/tip.2023.3326699}
}
Kostenlos starten →