Super-Resolution of Sentinel-2 Images: Learning a Globally Applicable Deep Neural Network
By Charis Lanaras, José M. Bioucas-Dias, Silvano Galliani, Emmanuel Baltsavias, Konrad Schindler. Isprs Journal of Photogrammetry and Remote Sensing, 2018
342 citations recorded by the source corpus.
Abstract The Sentinel-2 satellite mission delivers multi-spectral imagery with 13 spectral bands, acquired at three different spatial resolutions. The aim of this research is to super-resolve the lower-resolution (20 m and 60 m Ground Sampling Distance – GSD) bands to 10 m GSD, so as to obtain a complete data cube at the maximal sensor resolution. We employ a state-of-the-art convolutional neural network (CNN) to perform end-to-end upsampling, which is trained with data at lower resolution, i.e., from 40 → 20 m, respectively 360 → 60 m GSD. In this way, one has access to a virtually infinite amount of training data, by downsampling real Sentinel-2 images. We use data sampled globally over a wide range of geographical locations, to obtain a network that generalises across different climate zones and land-cover types, and can super-resolve arbitrary Sentinel-2 images without the need of retraining. In quantitative evaluations (at lower scale, where ground truth is available), our network, which we call DSen2, outperforms the best competing approach by almost 50% in RMSE, while better preserving the spectral characteristics. It also delivers visually convincing results at the full 10 m GSD.
Subfields: Computer Vision and Pattern Recognition, Media Technology
View Paper: https://www.semanticscholar.org/paper/3e4de68c121b9d7d5171a6b2462a4a188c8c1376
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Patrick Ebel
- Michael Schmitt
- Daifeng Peng
- Matteo Matteucci
- Yongjun Zhang
- Lorenzo Bruzzone
- Sara Comai
- Xu Huang
In Media Technology
- Yongjun Zhang
- Daifeng Peng
- Haiyan Guan
- Lorenzo Bruzzone
- Xu Huang
- Haiyong Ding
- Michael Schmitt
- Matteo Matteucci
Citations in the corpus
- Cloud removal in Sentinel-2 imagery using a deep residual neural network and SAR-optical data fusion (2020, positive): “Compromising between representational capacity and computational complexity, as well as considering own experiments and the reported experiences in Lanaras et al. (2018), Lim et al. (2017), residual units with = F 256 features were selected as a baseline for the DSen2-CR architecture.”
- SelfS2: Self-Supervised Transfer Learning for Sentinel-2 Multispectral Image Super-Resolution (2023, methodology): “Lanaras et al. [38] directly learn a deep convolutional neural network (CNN), which can be viewed as the inverse mapping of the observation model mentioned above, with the training data synthesized based on the scale invariance assumption.”
- SemiCDNet: A Semisupervised Convolutional Neural Network for Change Detection in High Resolution Remote-Sensing Images (2020, methodology): “Meanwhile, DL methods have also been widely used to solve RS problems such as image classification [21], object detection [22], [23], image super-resolution and denoising [24], [25], scene interpretation and segmentation [26].”
- Deep Learning for Land Use and Land Cover Classification Based on Hyperspectral and Multispectral Earth Observation Data: A Review (2020, background): “Another approach targeting the inner multi-modality of MSI data is by [164], wherein the proposed model super-resolves lower resolution spectral bands of Sentinel-2 data, using its higher resolution spectral bands.”