This project aims to develop an EU-wide data-driven solution for near real-time crop mapping by capitalizing on Earth Observation (EO) and diverse annotated training datasets. Currently leading a national Use Case (UC) for crop mapping, we seek to expand this capability to the European scale. The project will address the lack of homogenous, Europe-wide reference/training data by gathering and integrating annotated EO-CropType datasets from multiple countries. By pooling diverse "ground-truth" data, we aim to train robust AI models capable of handling the heterogeneity of European agriculture. This directly addresses the call's objective to enable the application of AI techniques and generate high-quality reference data sets. The project will ensure these datasets are FAIR (Findable, Accessible, Interoperable, Reusable) and will demonstrate the value of cross-border data sharing for AI model development.
We are looking to form a consortium of at least 3 partners from 3 different countries. We specifically seek:
1. Data Providers: Partners possessing national or regional annotated crop-type datasets (e.g., Land Parcel Identification Systems (LPIS) data, field surveys, or research trial data) who are willing to collaborate on data integration.
2. AI/ML Experts: Partners with expertise in processing large-scale EO time-series data and developing transfer learning models that can generalize across different agro-climatic zones.
3. End-Users: Public administrations or agricultural cooperatives interested in testing the resulting crop maps for policy monitoring or farm management support.

The Agricultural Research Organisation (ARO) - Volcani Institute is the research arm of the Israeli Ministry of Agriculture and Rural Development. We are a public research institution and a listed beneficiary of the Agriculture of Data partnership.
Our Expertise: We specialize in data-driven agriculture with a strong focus on Earth Observation (EO) and Artificial Intelligence (AI). We are currently leading a national-scale Use Case focused on near real-time crop mapping using satellite imagery and diverse ground-truth datasets. Our team possesses deep expertise in processing high-resolution EO time-series data and developing machine learning pipelines for agricultural monitoring.
Role in the Consortium: As a research performing organisation, we aim to act as a Use Case Provider and Technical Partner. We offer:
1. A Concrete Use Case: An operational national crop-mapping pilot ready for upscaling.
2. Reference Datasets: Access to annotated crop-type datasets to support the creation of EU-wide training sets for robust AI models.
3. Methodological Expertise: Experience in bridging the gap between remote sensing data and agronomic decision-support systems.