Potential of Convolutional Neural Networks for Earthquake‐triggered Mass Movement Detection Using Optical and SAR Data

Publikation: KonferenzbeitragPoster

Abstract

While Convolutional Neural Networks (CNNs) have reached good accuracies for object recognition in aerial images, only a few studies exist that use deep-learning methods and CNNs for landslide detection. The impact of using different input training data on landslide detection by the CNNs is unclear.
OriginalspracheEnglisch
PublikationsstatusVeröffentlicht - 12 Mai 2019
VeranstaltungESA Living Planet Symposium 2019 - Milan, Italien
Dauer: 13 Mai 201917 Mai 2019

Konferenz

KonferenzESA Living Planet Symposium 2019
LandItalien
OrtMilan
Zeitraum13/05/1917/05/19

Systematik der Wissenschaftszweige 2012

  • 105 Geowissenschaften

Zitieren