Image super-resolution with dense-sampling residual channel-spatial attention networks for multi-temporal remote sensing image classification
View / Open Files
Publication Date
2021-12-15Journal Title
International Journal of Applied Earth Observation and Geoinformation
ISSN
0303-2434
Publisher
Elsevier BV
Volume
104
Number
ARTN 102543
Type
Article
This Version
VoR
Metadata
Show full item recordCitation
Zhu, Y., Geiß, C., & So, E. (2021). Image super-resolution with dense-sampling residual channel-spatial attention networks for multi-temporal remote sensing image classification. International Journal of Applied Earth Observation and Geoinformation, 104 (ARTN 102543) https://doi.org/10.1016/j.jag.2021.102543
Abstract
Image super-resolution (SR) techniques can benefit a wide range of applications in the remote sensing (RS) community, including image classification. This issue is particularly relevant for image classification on time series data, considering RS datasets that feature long temporal coverage generally have a limited spatial resolution. Recent advances in deep learning brought new opportunities for enhancing the spatial resolution of
historic RS data. Numerous convolutional neural network (CNN)-based methods showed superior performance in terms of developing efficient end-to-end SR models for natural images. However, such models were rarely
exploited for promoting image classification based on multispectral RS data. This paper proposes a novel CNNbased framework to enhance the spatial resolution of time series multispectral RS images. Thereby, the proposed
SR model employs Residual Channel Attention Networks (RCAN) as a backbone structure, whereas based on this structure the proposed models uniquely integrate tailored channel-spatial attention and dense-sampling mechanisms for performance improvement. Subsequently, state-of-the-art CNN-based classifiers are incorporated to produce classification maps based on the enhanced time series data. The experiments proved that the proposed SR model can enable unambiguously better performance compared to RCAN and other (deep learning-based) SR techniques, especially in a domain adaptation context, i.e., leveraging Sentinel-2 images for generating SR Landsat images. Furthermore, the experimental results confirmed that the enhanced multi-temporal RS images can bring substantial improvement on fine-grained multi-temporal land use classification.
Keywords
Image super-resolution, Convolutional neural networks, Attention mechanism, Dense connection, Multi-temporal land use classification
Identifiers
External DOI: https://doi.org/10.1016/j.jag.2021.102543
This record's URL: https://www.repository.cam.ac.uk/handle/1810/329930
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International
Licence URL: https://creativecommons.org/licenses/by-nc-nd/4.0/
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.