Lightweight Spectral-Spatial Feature Extraction Network Based on Domain Generalization for Cross-Scene Hyperspectral Image Classification

The classification of land cover material based on hyperspectral image (HSI) has important research significance. Owing to the high cost of obtaining labeled samples and insufficient training samples, the research of cross-scene HSI classification (CS-HSIC) is receiving more and more attention. At p...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing 2024, Vol.62, p.1-14
Main Authors: Cui, Ying, Zhu, Longyu, Zhao, Chunhui, Wang, Liguo, Gao, Shan
Format: Article
Language:English
Subjects:
Attention mechanism
cascaded bottleneck
Classification
Convolution
Correlation
cross scene
Data mining
domain generalization (DG)
Feature extraction
hyperspectral image (HSI) classification
Hyperspectral imaging
Image classification
Land cover
Lightweight
Parameters
Performance enhancement
Task analysis
Training
Weight reduction
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The classification of land cover material based on hyperspectral image (HSI) has important research significance. Owing to the high cost of obtaining labeled samples and insufficient training samples, the research of cross-scene HSI classification (CS-HSIC) is receiving more and more attention. At present, the performance of the feature extraction module of CS-HSIC is relatively poor, and the number of training parameters is usually large. To fill the shortcomings of domain generalization (DG) methods and reduce the number of parameters, we propose a lightweight DG network with an attention-assisted cascaded bottleneck (ACB), and it adopts a lightweight bottleneck and multiattention design. This model is adept at extracting domain invariant information contained in the source domain (SD), and it may be flexibly embedded into other models. The experimental results show that our network has good classification accuracy and DG ability when the number of training samples is a little small. As a feature extraction subnetwork, it can improve the performance of the original model or reduce the required resources. The code will be available at https://github.com/zhulongyu1234/ACB .
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2024.3427375