Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/25326
Title: | Dynamic Convolution Self-Attention Network for Land-Cover Classification in VHR Remote-Sensing Images |
Authors: | Wang, X Zhang, Y Lei, T Wang, Y Zhai, Y Nandi, A |
Keywords: | Land-cover classification;feature fusion;self-attention;lightweight |
Issue Date: | 3-Oct-2022 |
Publisher: | MDPI |
Citation: | Wang, X. et al (2022) 'Dynamic Convolution Self-Attention Network for Land-Cover Classification in VHR Remote-Sensing Images'. Remote Sensing, 14 (19), 4941, pp.1 - 20. https://doi.org/10.3390/rs14194941 |
Abstract: | Copyright © 2022 by the authors. The current deep convolutional neural networks for very-high-resolution (VHR) remote-sensing image land-cover classification often suffer from two challenges. First, the feature maps extracted by network encoders based on vanilla convolution usually contain a lot of redundant information, which easily causes misclassification of land cover. Moreover, these encoders usually require a large number of parameters and high computational costs. Second, as remote-sensing images are complex and contain many objects with large-scale variances, it is difficult to use the popular feature fusion modules to improve the representation ability of networks. To address the above issues, we propose a dynamic convolution self-attention network (DCSA-Net) for VHR remote-sensing image land-cover classification. The proposed network has two advantages. On one hand, we designed a lightweight dynamic convolution module (LDCM) by using dynamic convolution and a self-attention mechanism. This module can extract more useful image features than vanilla convolution, avoiding the negative effect of useless feature maps on land-cover classification. On the other hand, we designed a context information aggregation module (CIAM) with a ladder structure to enlarge the receptive field. This module can aggregate multi-scale contexture information from feature maps with different resolutions using a dense connection. Experiment results show that the proposed DCSA-Net is superior to state-of-the-art networks due to higher accuracy of land-cover classification, fewer parameters, and lower computational cost. The source code is made public available. |
Description: | Data Availability Statement: The datasets used in this study have been published, and their addresses are https://www2.isprs.org/commissions/comm2/wg4/benchmark/2d-sem-label-potsdam/ (accessed on 30 January 2021) and https://www2.isprs.org/commissions/comm2/wg4/benchmark/2d-sem-label-vaihingen/ (accessed on 30 January 2021). |
URI: | https://bura.brunel.ac.uk/handle/2438/25326 |
DOI: | https://doi.org/10.3390/rs14194941 |
Other Identifiers: | ORCID iDs: Xuan Wang https://orcid.org/0000-0002-0842-6511; Tao Lei https://orcid.org/0000-0002-2104-9298; Yingbo Wang https://orcid.org/0000-0001-6447-8730; Asoke K. Nandi https://orcid.org/0000-0001-6248-2875. 4941 |
Appears in Collections: | Dept of Electronic and Electrical Engineering Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
FullText.pdf | Copyright © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). | 4.82 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License