Hyperspectral Image Classification Based on Multi-stage Vision Transformer with Stacked Samples

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)


Hyperspectral image classification (HSIC) is a task assigning the correct label to each pixel. It is a hot topic in the remote sensing field, which has been processed in several deep learning methods. Recently, there are some works that apply Vision Transformer (ViT) methods to the HSIC task, but the performance is not as good as some CNN-structured methods, considering that Vision Transformer uses attention to capture global information but ignores local characteristics. In this paper, a multi-stage Vision Transformer model referring to the feature extraction structure of CNN is proposed, and the result shows the realizability and reliability. Besides, experiments show that the modified ViT structure needs more samples for training. An innovative data augmentation method is used to generate extended samples with virtual yet reliable labels. The generated samples are combined with the original ones as the stacked samples, which are used for the following feature extraction process. Experiments explain the optimization of the multi-stage Vision Transformer structure with stacked samples in the accuracy term compared with other methods.

Original languageEnglish
Title of host publicationTENCON 2021 - 2021 IEEE Region 10 Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages6
ISBN (Electronic)9781665495325
Publication statusPublished - 2021
Event2021 IEEE Region 10 Conference, TENCON 2021 - Auckland, New Zealand
Duration: 2021 Dec 72021 Dec 10

Publication series

NameIEEE Region 10 Annual International Conference, Proceedings/TENCON
ISSN (Print)2159-3442
ISSN (Electronic)2159-3450


Conference2021 IEEE Region 10 Conference, TENCON 2021
Country/TerritoryNew Zealand


  • Hyperspectral image classification
  • Vision Transformer
  • data augmentation
  • deep learning
  • image processing

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Hyperspectral Image Classification Based on Multi-stage Vision Transformer with Stacked Samples'. Together they form a unique fingerprint.

Cite this