A Transformer-Unet Generative Adversarial Network for the Super-Resolution Reconstruction of DEMs

A new model called the Transformer-Unet Generative Adversarial Network (TUGAN) is proposed for super-resolution reconstruction of digital elevation models (DEMs). Digital elevation models are used in many fields, including environmental science, geology and agriculture. The proposed model uses a sel...

Full description

Saved in:
Bibliographic Details
Published in:Remote sensing (Basel, Switzerland) Switzerland), 2024-10, Vol.16 (19), p.3676
Main Authors: Zheng, Xin, Xu, Zhaoqi, Yin, Qian, Bao, Zelun, Chen, Zhirui, Wang, Sizhu
Format: Article
Language:English
Subjects:
Algorithms
Artificial intelligence
Comparative analysis
Deep learning
DEM
Digital Elevation Models
Discriminators
Electronic data processing
Elevation
Environmental science
Feedback
GAN
Generative adversarial networks
Geology
Machine learning
Methods
Multilayers
Neural networks
Reconstruction
Self-similarity
super-resolution
Technology application
Topography
transformer
Transformers
U-Net
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new model called the Transformer-Unet Generative Adversarial Network (TUGAN) is proposed for super-resolution reconstruction of digital elevation models (DEMs). Digital elevation models are used in many fields, including environmental science, geology and agriculture. The proposed model uses a self-similarity Transformer (SSTrans) as the generator and U-Net as the discriminator. SSTrans, a model that we previously proposed, can yield good reconstruction results in structurally complex areas but has little advantage when the surface is simple and smooth because too many additional details have been added to the data. To resolve this issue, we propose the novel TUGAN model, where U-Net is capable of multilayer jump connections, which enables the discriminator to consider both global and local information when making judgments. The experiments show that TUGAN achieves state-of-the-art results for all types of terrain details.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs16193676