Super-resolution EO-based area monitoring markers computed over the Lithuanian pilot region (2022)

In the context of the EU-funded project DIONE (No. 870378), the following EO-based monitoring marker maps were released over defined pilot areas over the Lithuanian pilot country, enhanced by features raised from the Super-resolution models. It involved the implementation of matching marking and data fusion deep learning algorithms, which attempt to support the extraction of useful information from highly variable inputs. This will further allow the distinction of landscape features, which would otherwise not be available in initially acquired Sentinel-2 data. The use of VHR data  (Copernicus Contributing Missions) in combination with drone imagery will enhance the super-resolution modelling capabilities, enabling the augmentation of the training dataset (spatio-temporal scale) and subsequently leading to increased model performance.
The goal was to enhance the outputs of the area monitoring markers and especially in the monitoring of small (i.e. 100m2), narrow and elongated parcels. 

For the needs of DIONE, the aforementioned data were explored and the following area-based monitoring markers were calculated from 01-01-2022 until 01-08-2022 providing tailored information for the needs of the National Paying Agency of Lithuania. 

Mowing marker: used to detect mowing events on meadow/grass like Features Of Interest (FOI)

Similarity and distance markers: used to give additional context to the crop classification and to detect erroneous claims

Crop-type marker: used to detect the specific crop growing on the FOI

This dataset is comprised of one geopackage file, the "S2SR-study-geopackage.gpkg", which was computed for the Lithuanian pilot region. Descriptions are given below.

Super-resolution Markers dataset: Markers were computed for 16872 FOIs that contain less than 1 Sentinel-2 pixel, using signals from 2022-01-01 until 2022-08-01. 

Description of the information contained in the corresponding "Super-Resolution markers" dataset

Attribute name 
Description 

CROP_LABEL
Reference ID of the polygon 

POLY_ID 
Declared crop group

crop_group_prediction_1_classification
The FOI label as predicted by the crop group (v2) model

crop_group_prediction_1_classification_score
The pseudoprobability of the crop-group (v1) prediction. A score close to 1 indicates that the model is very confident in the prediction

crop_group_prediction_2_classification
The FOI label as predicted by the crop group (v2) model

crop_group_prediction_2_classification_score
The pseudoprobability of the crop-group (v2) prediction. A score close to 1 indicates that the model is very confident in the prediction

distance_classification
Most similar crops according to the distance marker

distance_classification_score
Distance marker score of a FOI when compared to nearby FOIs with the same claim. A value close to 100 indicates that a FOI is not similar to other FOIs with the same claim.

mowing_event_count
Number of detected mowing events in the observation period

similarity_classification
Most similar crops according to similarity marker

similarity_classification_score
Similarity marker score of a FOI when compared to nearby FOIs with the same claim. A value close to 100 indicates that a FOI is not similar to other FOIs with the same claim