Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/132006
Title: Analysis of the effects of applying a restoration process to despeckled SAR imagery
Authors: Vasquez Salazar, Ruben Dario 
Cardona Mesa, Ahmed Alejandro 
Gomez, Luis 
Travieso-González, Carlos M. 
Garavito-Gonzalez, Andres F.
Vasquez-Cano, Esteban
UNESCO Clasification: 33 Ciencias tecnológicas
Keywords: Autoencoder
Deep Learning
Despeckling Filter
Synthetic Aperture Radar
Transformer
Issue Date: 2024
Conference: International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing 2024 
Abstract: Despeckling SAR images constitute an important step when utilizing these types of images to ensure the quality of applications reliant upon them. After applying any despeckling filter, there will be an effect on the quality of the images, including edges, details, or possibly some undesired artifacts. In this paper, a restoration process using the SwinIR transformer is applied over-filtered images to reconstruct or recover some of their lost details. Different metrics are applied to filtered and reconstructed images to measure the impact and benefits of this process.
URI: http://hdl.handle.net/10553/132006
ISBN: 9798350389678
DOI: 10.1109/MIGARS61408.2024.10544393
Source: 2024 International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing, MIGARS 2024[EISSN ], (Enero 2024)
Appears in Collections:Actas de congresos
Show full item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.