Journal article · 2023

Super Resolution Approach with Convolutional Autoencoder Neural Network for Sentinel-2 Satellite Imagery

Turkish titleSentinel-2 Uydu Görüntüleri için Evrişimli Otokodlayıcı Sinir Ağı ile Süper Çözünürlük Yaklaşımı

Ahmet Ertuğrul ArıkORCID 0000-0002-7952-4311 (external link)Reha PaşaoğluNuri Emrahaoğlu

Turkish Journal of Remote Sensing and GIS, 4(2), 231–241

Year
2023
Publication date
Journal or conference
Turkish Journal of Remote Sensing and GIS
Volume
4
Issue
2
Pages
231–241
DOI
10.48123/rsgis.1254716 (external link)

Abstract

The article presents SEN-2_CAENET, a convolutional autoencoder-based deep learning model for increasing the spatial resolution of Sentinel-2 satellite imagery. The model is evaluated using PSNR, MSE and SSIM metrics. The reported tests show that SEN-2_CAENET achieved stronger results than the SRCNN baseline used for comparison.

Keywords

  • Artificial neural networks
  • Autoencoders
  • Super resolution
  • Image processing
  • Remote sensing

Core method

A convolutional autoencoder architecture named SEN-2_CAENET, evaluated with PSNR, MSE and SSIM against SRCNN.

Data used

Sentinel-2 satellite imagery prepared for super-resolution model training and evaluation.

Verified key results

  • SEN-2_CAENET outperformed SRCNN in the reported PSNR, MSE and SSIM evaluation.

APA citation

Arık, A. E., Paşaoğlu, R., & Emrahaoğlu, N. (2023). Sentinel-2 Uydu Görüntüleri için Evrişimli Otokodlayıcı Sinir Ağı ile Süper Çözünürlük Yaklaşımı. Türk Uzaktan Algılama ve CBS Dergisi, 4(2), 231-241. https://doi.org/10.48123/rsgis.1254716

BibTeX

@article{ArikPasogluEmrahoglu2023SuperResolution,
  author  = {Arık, Ahmet Ertuğrul and Paşaoğlu, Reha and Emrahaoğlu, Nuri},
  title   = {Sentinel-2 Uydu Görüntüleri için Evrişimli Otokodlayıcı Sinir Ağı ile Süper Çözünürlük Yaklaşımı},
  journal = {Türk Uzaktan Algılama ve CBS Dergisi},
  year    = {2023},
  volume  = {4},
  number  = {2},
  pages   = {231--241},
  doi     = {10.48123/rsgis.1254716}
}