Document Type
Conference Object
Disciplines
1.5 EARTH AND RELATED ENVIRONMENTAL SCIENCES, Environmental sciences, *human – machine relations, 5.2 ECONOMICS AND BUSINESS
Abstract
High spatial resolution satellite imagery is essential for environmental monitoring, forest disturbance assessment, and precision agriculture. However, widely available sensors such as Sentinel-2 provide only moderate spatial resolution, limiting the detection of fine spatial structures and textures. Recent advances in generative artificial intelligence, particularly diffusion models, have demonstrated strong performance in image restoration and super-resolution tasks. Despite this progress, preserving directional structures and geometric patterns remains a major challenge in remote sensing imagery. In this work, we propose a super-resolution framework that integrates the Contourlet Transform with Latent Diffusion Models (LDMs). The contourlet transform extracts multi-scale directional features that capture edges and textures, which are used to guide the latent diffusion model during image reconstruction
DOI
https://doi.org/10.21427/7NTB-AD34
Recommended Citation
Amro, Israa and Wall, P.J., "Contourlet-Guided Latent Diffusion for Super-Resolution of Sentinel-2 Imagery" (2026). Other. 1.
https://arrow.tudublin.ie/adaptother/1
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Publication Details
This research was supported by the ADAPT Research Centre and the MSCA Cofund DIGI+ programme under the European Union’s Marie Sk lodowska-Curie Actions.