User: Guest  Login
Title:

Self-attention for raw optical Satellite Time Series Classification

Document type:
Forschungsdaten
Publication date:
01.07.2021
Responsible:
Körner, Marco
Authors:
Rußwurm, Marc ; Körner, Marco
Author affiliation:
TUM
Publisher:
TUM
End date of data production:
07.04.2020
Subject area:
BAU Bauingenieurwesen, Vermessungswesen; DAT Datenverarbeitung, Informatik; GEO Geowissenschaften
Resource type:
Abbildungen von Objekten / image of objects
Other resource types:
Zipped Image and Time Series Data
Data type:
Bilder / images; mehrdimensionale Visualisierungen oder Modelle / models; Tabellen / tables
Description:
Dataset for crop type mapping in Bavaria. Contains raw (unpreprocessed) Sentinel 2 satellite time series acquired over the year 2018 associated with crop type labels of the field parcels. This dataset was used to compare several deep learning models on preprocessed satellite data to industry standards with raw satellite time series to assess the robustness of the tested deep learning models to noise in the data.
Method of data assessment:
Download from Sentinel 2 satellite data archive, image processing (cropping to relevant field parcels). Association with crop type labels acquired from the Bavarian Ministry of Agriculture
Links:

Link Artikel: https://www.sciencedirect.com/science/article/abs/pii/S0924271620301647 and https://mediatum.ub.tum.de/1602502
DOI: 10.1016/j.isprsjprs.2020.06.006
Code Repository: https://github.com/MarcCoru/crop-type-mapping

Key words:
Satellite Time Series analysis; Multi-temporal image analysis; Vegetation classification; Crop Type Mapping; Deep Learning; Machine Learning
Technical remarks:
View and download (2.66 GB, 6 files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1612845):
rsync rsync://m1612845@dataserv.ub.tum.de/m1612845/
Language:
en
Rights:
by, http://creativecommons.org/licenses/by/4.0
 BibTeX