Cristin-resultat-ID: 1696634
Sist endret: 5. februar 2020 16:05
NVI-rapporteringsår: 2019
Resultat
Vitenskapelig artikkel
2019

Void Filling of Digital Elevation Models With Deep Generative Models

Bidragsytere:
  • Konstantinos Gavriil
  • Agnar Georg Peder Muntingh og
  • Oliver Joseph David Barrowclough

Tidsskrift

IEEE Geoscience and Remote Sensing Letters
ISSN 1545-598X
e-ISSN 1558-0571
NVI-nivå 1

Om resultatet

Vitenskapelig artikkel
Publiseringsår: 2019
Volum: 16
Hefte: 10
Sider: 1645 - 1649

Importkilder

Scopus-ID: 2-s2.0-85076575363

Klassifisering

Vitenskapsdisipliner

Matematikk og naturvitenskap

Emneord

Digitale høydemodeller • Fjernanalyse

Beskrivelse Beskrivelse

Tittel

Void Filling of Digital Elevation Models With Deep Generative Models

Sammendrag

In recent years, advances in machine learning algorithms, cheap computational resources, and the availability of big data have spurred the deep learning revolution in various application domains. In particular, supervised learning techniques in image analysis have led to a superhuman performance in various tasks, such as classification, localization, and segmentation, whereas unsupervised learning techniques based on increasingly advanced generative models have been applied to generate high-resolution synthetic images indistinguishable from real images. In this letter, we consider a state-of-the-art machine learning model for image inpainting, namely, a Wasserstein Generative Adversarial Network based on a fully convolutional architecture with a contextual attention mechanism. We show that this model can be successfully transferred to the setting of digital elevation models for the purpose of generating semantically plausible data for filling voids. Training, testing, and experimentation are done on GeoTIFF data from various regions in Norway, made openly available by the Norwegian Mapping Authority.

Bidragsytere

Konstantinos Gavriil

  • Tilknyttet:
    Forfatter
    ved Technische Universität Wien

Agnar Georg Peder Muntingh

  • Tilknyttet:
    Forfatter
    ved Mathematics and Cybernetics ved SINTEF AS

Oliver Joseph David Barrowclough

  • Tilknyttet:
    Forfatter
    ved Mathematics and Cybernetics ved SINTEF AS
1 - 3 av 3