The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B1-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B1-2020, 9–14, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-9-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B1-2020, 9–14, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-9-2020

  06 Aug 2020

06 Aug 2020

A GENERATIVE ADVERSARIAL NETWORK APPROACH FOR SUPER-RESOLUTION OF SENTINEL-2 SATELLITE IMAGES

F. Pineda1,2, V. Ayma1, and C. Beltran1 F. Pineda et al.
  • 1Pontificia Universidad Católica del Perú, Department of Engineering, Artificial Intelligence Research Group (IA-PUCP), Lima, Perú
  • 2Universidad Nacional del Altiplano, Puno, Perú

Keywords: PeruSat-1, Sentinel-2, Super-Resolution, GAN

Abstract. High-resolution satellite images have always been in high demand due to the greater detail and precision they offer, as well as the wide scope of the fields in which they could be applied; however, satellites in operation offering very high-resolution (VHR) images has experienced an important increase, but they remain as a smaller proportion against existing lower resolution (HR) satellites. Recent models of convolutional neural networks (CNN) are very suitable for applications with image processing, like resolution enhancement of images; but in order to obtain an acceptable result, it is important, not only to define the kind of CNN architecture but the reference set of images to train the model. Our work proposes an alternative to improve the spatial resolution of HR images obtained by Sentinel-2 satellite by using the VHR images from PeruSat1, a peruvian satellite, which serve as the reference for the super-resolution approach implementation based on a Generative Adversarial Network (GAN) model, as an alternative for obtaining VHR images. The VHR PeruSat-1 image dataset is used for the training process of the network. The results obtained were analyzed considering the Peak Signal to Noise Ratios (PSNR) and the Structural Similarity (SSIM). Finally, some visual outcomes, over a given testing dataset, are presented so the performance of the model could be analyzed as well.