The aim of the DIONE project was to propose a close-to-market area-based direct payments monitoring toolbox that would enable more frequent, accurate and inexpensive compliance checks.
To achieve this, the project has conducted work to:
- Integrate generated crop-type maps to enable their direct use by the paying agencies;
- Include in the analysis the so far neglected ecological focus area types (fallow land of all sizes, buffer strips, hedges, trees) by making use of high-resolution technology that improves the 10-20m Sentinel resolution to a 5-10m range. This is enabled through Machine-Learning (ML) based post-processing and data fusion of Copernicus DIAS-sourced data with targeted drone-obtained data.
- Complement the use of Earth Observation data with a system of reliable, ground-based geo-tagged photos, captured by the farmers that exploits (i) advances that allow for improved positional accuracy, (ii) low-footprint encryption techniques for improved data security and reliability and (iii) image detecting manipulation techniques (image forensics).
- Implement a Green Compliance toolbox, integrated with the paying agencies’ aforementioned tools. This benefits from (i) low-cost spectral sensors measuring soil quality and assessing the status of land-degradation in the land parcels and (ii) an ML-based inferencing system deployed on a larger scale (regional, national) to quantify the levels of some of the monitored parameters and consequently extract tangible environmental performance metrics for an entire region.
