Next Article in Journal
Deep Compressed Communication and Application in Multi-Robot 2D-Lidar SLAM: An Intelligent Huffman Algorithm
Previous Article in Journal
Ammonia Detection by Electronic Noses for a Safer Work Environment
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

DMAF-NET: Deep Multi-Scale Attention Fusion Network for Hyperspectral Image Classification with Limited Samples

by
Hufeng Guo
1,2 and
Wenyi Liu
1,*
1
State Key Laboratory of Dynamic Measurement Technology, School of Instrument and Electronics, North University of China, Tai Yuan 030051, China
2
Department of Transportation Information Engineering, Henan College of Transportation, Zhengzhou 451460, China
*
Author to whom correspondence should be addressed.
Sensors 2024, 24(10), 3153; https://doi.org/10.3390/s24103153
Submission received: 9 April 2024 / Revised: 6 May 2024 / Accepted: 13 May 2024 / Published: 15 May 2024
(This article belongs to the Special Issue Remote Sensing Technology for Agricultural and Land Management)

Abstract

In recent years, deep learning methods have achieved remarkable success in hyperspectral image classification (HSIC), and the utilization of convolutional neural networks (CNNs) has proven to be highly effective. However, there are still several critical issues that need to be addressed in the HSIC task, such as the lack of labeled training samples, which constrains the classification accuracy and generalization ability of CNNs. To address this problem, a deep multi-scale attention fusion network (DMAF-NET) is proposed in this paper. This network is based on multi-scale features and fully exploits the deep features of samples from multiple levels and different perspectives with an aim to enhance HSIC results using limited samples. The innovation of this article is mainly reflected in three aspects: Firstly, a novel baseline network for multi-scale feature extraction is designed with a pyramid structure and densely connected 3D octave convolutional network enabling the extraction of deep-level information from features at different granularities. Secondly, a multi-scale spatial–spectral attention module and a pyramidal multi-scale channel attention module are designed, respectively. This allows modeling of the comprehensive dependencies of coordinates and directions, local and global, in four dimensions. Finally, a multi-attention fusion module is designed to effectively combine feature mappings extracted from multiple branches. Extensive experiments on four popular datasets demonstrate that the proposed method can achieve high classification accuracy even with fewer labeled samples.
Keywords: convolutional neural network (CNN); hyperspectral image (HSI) classification; limited samples; multi-scale feature extraction; multi-scale spatial–spectral attention; pyramidal multi-scale channel attention; multi-attention feature fusion convolutional neural network (CNN); hyperspectral image (HSI) classification; limited samples; multi-scale feature extraction; multi-scale spatial–spectral attention; pyramidal multi-scale channel attention; multi-attention feature fusion

Share and Cite

MDPI and ACS Style

Guo, H.; Liu, W. DMAF-NET: Deep Multi-Scale Attention Fusion Network for Hyperspectral Image Classification with Limited Samples. Sensors 2024, 24, 3153. https://doi.org/10.3390/s24103153

AMA Style

Guo H, Liu W. DMAF-NET: Deep Multi-Scale Attention Fusion Network for Hyperspectral Image Classification with Limited Samples. Sensors. 2024; 24(10):3153. https://doi.org/10.3390/s24103153

Chicago/Turabian Style

Guo, Hufeng, and Wenyi Liu. 2024. "DMAF-NET: Deep Multi-Scale Attention Fusion Network for Hyperspectral Image Classification with Limited Samples" Sensors 24, no. 10: 3153. https://doi.org/10.3390/s24103153

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop