EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Scale Spatial–Spectral Residual Attention Network for Hyperspectral Image Classification

Qinggang Wu, Mengkun He, Zhongchi Liu, Yanyan Liu

Electronics · 05.01.2024 · 13 (2) · 262

DOI: 10.3390/electronics13020262 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 05.01.2024. Verfasst von Qinggang Wu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation, Young Backbone Teacher Training Program of Henan Province, Scientific and Technological Research Project of Henan Provincial Department of Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral images (HSIs) encapsulate a vast amount of information due to their expansive size and high number of channel dimensions. However, they are insufficiently utilized for ineffective feature extraction, particularly for regions with few samples and predominant edges. To fully leverage the spatial–spectral features of HSIs, a dual-branch multi-scale spatial–spectral residual attention network (MSRAN) that integrates multi-scale feature extraction with residual attention mechanisms is proposed. MSRAN independently extracts spatial and spectral features through dual branches, minimizing the interference between these features and enhancing the focus on feature extraction in different dimensions. Specifically, in the spectral feature extraction branch, diverse-scale 3D convolution kernels capture extended spectral sequence characteristics and neighborhood spectral features. The convolution fusion emphasizes the weight of the central pixel to be classified, followed by the use of spectral residual attention mechanisms to extract enhanced central-pixel spectral features. In the spatial feature extraction branch, multi-level receptive fields are utilized to extract various fine-grained spatial contours, edges, and local detailed features, which are further processed through spatial residual attention to effectively extract spatial composite features. Finally, the convolution fusion module adaptively integrates the center-enhanced spectral features with multi-level fine-grained spatial features for classification. Extensive comparative experiments and ablation studies demonstrate that MSRAN achieves highly competitive results on two classic datasets from Pavia University and Salinas as well as on a novel dataset of WHU-Hi-LongKou.

Förderung

  • National Natural Science Foundation
  • Young Backbone Teacher Training Program of Henan Province
  • Scientific and Technological Research Project of Henan Provincial Department of Science and Technology

Zitieren

APA

Wu, Q., He, M., Liu, Z., & Liu, Y. (2024). Multi-Scale Spatial–Spectral Residual Attention Network for Hyperspectral Image Classification. Electronics, 13(2), 262. https://doi.org/10.3390/electronics13020262

Vancouver

Wu Q, He M, Liu Z, Liu Y. Multi-Scale Spatial–Spectral Residual Attention Network for Hyperspectral Image Classification. Electronics. 2024;13(2):262. doi:10.3390/electronics13020262

BibTeX
@article{Wu2024,
  author = {Wu, Qinggang and He, Mengkun and Liu, Zhongchi and Liu, Yanyan},
  title = {Multi-Scale Spatial–Spectral Residual Attention Network for Hyperspectral Image Classification},
  journal = {Electronics},
  year = {2024},
  volume = {13},
  number = {2},
  pages = {262},
  doi = {10.3390/electronics13020262}
}
Kostenlos starten →