EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hyperspectral Image Classification Using Spectral–Spatial Double-Branch Attention Mechanism

Jianfang Kang, Yaonan Zhang, Xinchao Liu, Zhongxin Cheng

Remote Sensing · 02.01.2024 · 16 (1) · 193

DOI: 10.3390/rs16010193 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 02.01.2024. Verfasst von Jianfang Kang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 43-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key R&D Program of China, National Cryosphere Desert Data Center, Research and Development in Artificial Intelligence Data Fusion.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent years, deep learning methods utilizing convolutional neural networks have been extensively employed in hyperspectral image classification (HSI) applications. Nevertheless, while a substantial number of stacked 3D convolutions can indeed achieve high classification accuracy, they also introduce a significant number of parameters to the model, resulting in inefficiency. Furthermore, such intricate models often exhibit limited classification accuracy when confronted with restricted sample data, i.e., small sample problems. Therefore, we propose a spectral–spatial double-branch network (SSDBN) with an attention mechanism for HSI classification. The SSDBN is designed with two independent branches to extract spectral and spatial features, respectively, incorporating multi-scale 2D convolution modules, long short-term memory (LSTM), and an attention mechanism. The flexible use of 2D convolution, instead of 3D convolution, significantly reduces the model’s parameter count, while the effective spectral–spatial double-branch feature extraction method allows SSDBN to perform exceptionally well in handling small sample problems. When tested on 5%, 0.5%, and 5% of the Indian Pines, Pavia University, and Kennedy Space Center datasets, SSDBN achieved classification accuracies of 97.56%, 96.85%, and 98.68%, respectively. Additionally, we conducted a comparison of training and testing times, with results demonstrating the remarkable efficiency of SSDBN.

Förderung

  • National Key R&D Program of China
  • National Cryosphere Desert Data Center
  • Research and Development in Artificial Intelligence Data Fusion
  • Demonstration of Digital Empowerment Applications

Zitieren

APA

Kang, J., Zhang, Y., Liu, X., & Cheng, Z. (2024). Hyperspectral Image Classification Using Spectral–Spatial Double-Branch Attention Mechanism. Remote Sensing, 16(1), 193. https://doi.org/10.3390/rs16010193

Vancouver

Kang J, Zhang Y, Liu X, Cheng Z. Hyperspectral Image Classification Using Spectral–Spatial Double-Branch Attention Mechanism. Remote Sensing. 2024;16(1):193. doi:10.3390/rs16010193

BibTeX
@article{Kang2024,
  author = {Kang, Jianfang and Zhang, Yaonan and Liu, Xinchao and Cheng, Zhongxin},
  title = {Hyperspectral Image Classification Using Spectral–Spatial Double-Branch Attention Mechanism},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {1},
  pages = {193},
  doi = {10.3390/rs16010193}
}
Kostenlos starten →