AgriSen-COG, a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping Using Deep Learning
Author:
Open Access
- 7 June 2023
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 15 (12) , 2980
Abstract
With the increasing volume of collected Earth observation (EO) data, artificial intelligence (AI) methods have become state-of-the-art in processing and analyzing them. However, there is still a lack of high-quality, large-scale EO datasets for training robust networks. This paper presents AgriSen-COG, a large-scale benchmark dataset for crop type mapping based on Sentinel-2 data. AgriSen-COG deals with the challenges of remote sensing (RS) datasets. First, it includes data from five different European countries (Austria, Belgium, Spain, Denmark, and the Netherlands), targeting the problem of domain adaptation. Second, it is multitemporal and multiyear (2019–2020), therefore enabling analysis based on the growth of crops in time and yearly variability. Third, AgriSen-COG includes an anomaly detection preprocessing step, which reduces the amount of mislabeled information. AgriSen-COG comprises 6,972,485 parcels, making it the most extensive available dataset for crop type mapping. It includes two types of data: pixel-level data and parcel aggregated information. By carrying this out, we target two computer vision (CV) problems: semantic segmentation and classification. To establish the validity of the proposed dataset, we conducted several experiments using state-of-the-art deep-learning models for temporal semantic segmentation with pixel-level data (U-Net and ConvStar networks) and time-series classification with parcel aggregated information (LSTM, Transformer, TempCNN networks). The most popular models (U-Net and LSTM) achieve the best performance in the Belgium region, with a weighted F1 score of 0.956 (U-Net) and 0.918 (LSTM).The proposed data are distributed as a cloud-optimized GeoTIFF (COG), together with a SpatioTemporal Asset Catalog (STAC), which makes AgriSen-COG a findable, accessible, interoperable, and reusable (FAIR) dataset.Keywords
Funding Information
- Romanian Ministry of Research and Innovation, CNCS—UEFISCDI (PN-III-P4-ID-PCE-2016-0842, PN III-P4-ID-PCE-2020-0407, 101003517, 4000125049/18/NL/CBI, POC/163/1/3/)
- PNCDI III, COCO (PN-III-P4-ID-PCE-2016-0842, PN III-P4-ID-PCE-2020-0407, 101003517, 4000125049/18/NL/CBI, POC/163/1/3/)
- EU Horizon 2020 research and innovation programme (PN-III-P4-ID-PCE-2016-0842, PN III-P4-ID-PCE-2020-0407, 101003517, 4000125049/18/NL/CBI, POC/163/1/3/)
- European Space Agency (PN-III-P4-ID-PCE-2016-0842, PN III-P4-ID-PCE-2020-0407, 101003517, 4000125049/18/NL/CBI, POC/163/1/3/)
- Romanian UEFISCD FUSE4DL (PN-III-P4-ID-PCE-2016-0842, PN III-P4-ID-PCE-2020-0407, 101003517, 4000125049/18/NL/CBI, POC/163/1/3/)
This publication has 59 references indexed in Scilit:
- xarray: N-D labeled Arrays and Datasets in PythonJournal of Open Research Software, 2017
- Practical Hadoop Ecosystem2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational ServicesRemote Sensing of Environment, 2012
- OGC standards and the space community — Processes, application and value2011
- A global averaging method for dynamic time warping, with applications to clusteringPattern Recognition, 2011
- Long Short-Term MemoryNeural Computation, 1997
- Serial Order: A Parallel Distributed Processing Approach1997
- A Threshold Selection Method from Gray-Level HistogramsIEEE Transactions on Systems, Man, and Cybernetics, 1979
- Nearest neighbor pattern classificationIEEE Transactions on Information Theory, 1967