A Self-Trained Model for Cloud, Shadow and Snow Detection in Sentinel-2 Images of Snow- and Ice-Covered Regions

Screening clouds, shadows, and snow is a critical pre-processing step in many remote-sensing data processing pipelines that operate on satellite image data from polar and high mountain regions. We observe that the results of the state-of-the-art Fmask algorithm are not very accurate in polar and hig...

Полное описание

Сохранить в:
Библиографические подробности
Главные авторы: Kamal Gopikrishnan Nambiar, Veniamin I. Morgenshtern, Philipp Hochreuther, Thorsten Seehaus, Matthias Holger Braun
Формат: Artigo
Язык:Inglês
Опубликовано: MDPI AG 2022-04-01
Серии:Remote Sensing
Предметы:
deep learning
semi-supervised learning
semantic segmentation
self-training
automatic cloud screening
Fmask
Online-ссылка:https://www.mdpi.com/2072-4292/14/8/1825
Метки: Добавить метку
Нет меток, Требуется 1-ая метка записи!

Схожие документы