Catalogue of Services

Our service is designed to assist clients in identifying and evaluating potential agricultural applications for AI or robotics technologies originally developed for other sectors. Clients can provide detailed information about their technology, including its current applications and technical specifications. Our expert team will conduct extensive research to identify possible use cases related to the agricultural and food production sector, analysing factors such as feasibility, potential benefits, and relevance to current agricultural challenges. The tools used for this work will depend on the use case and may include, for example, online research as well as expert interviews. The customer will receive the results as a comprehensive report outlining the identified potential applications, adoptions needed and an assessment of their relevance and impact within the agricultural sector. This service enables clients to explore new markets and adapt their technologies to meet the evolving needs of modern agriculture.

Desk assessment

Market research

Data analysis to optimise AI model training

Institut de l'elevage (IDELE)

Our service enhances your dataset's quality using descriptive statistics to gain insights and ensure robustness. We use measures like mean, median, standard deviation, etc., to identify central tendencies and variability. We perform plausibility filtering to check for logical consistency and outlier filtering to remove data points that could skew predictions. This careful data preparation refines your dataset, boosting the accuracy and reliability of your model training. Tailored to your needs, these methods optimise your dataset for better model performance.

Data analysis

Data and media labelling for AI training

This service specialises in accurately annotating and labelling datasets or multimedia content, including photos, videos, and audio, specifically for training artificial intelligence algorithms in the ruminant livestock sector. By meticulously tagging key features such as animal behaviour, physical conditions, environmental factors, and health indicators, we provide high-quality datasets that enhance the precision of AI models used in livestock monitoring, health assessment, and farm management solutions. Our service ensures that AI solutions for ruminant farming are trained with contextually rich, well-labelled data, enabling the development of advanced tools for animal welfare, productivity optimisation, and predictive analytics in the livestock industry.

Institut de l'elevage (IDELE)

Data and media labelling for AI training

Location

Remote

Tree Crops

This service specializes in accurately annotating and labeling datasets or multimedia content, including photos, videos, and audio, specifically for training Artificial Intelligence algorithms in the ruminant livestock sector. By meticulously tagging key features such as animal behavior, physical conditions, environmental factors, and health indicators, we provide high-quality datasets that enhance the precision of AI models used in livestock monitoring, health assessment, and farm management solutions.

Our service ensures that AI solutions for ruminant farming are trained with contextually rich, well-labeled data, enabling the development of advanced tools for animal welfare, productivity optimization, and predictive analytics in the livestock industry.

Data augmentation to optimise AI model training and prevent the model from overfitting

In preparing image or sensor data, pre-processing is crucial. We apply techniques like normalisation for data uniformity and correlation analysis to uncover patterns, enhancing interpretability. Dimension reduction optimises computational efficiency, while adding randomised noise and artificial expansion improves model generalisation. We implement anonymisation to protect data privacy and ensure precise data labelling for supervised learning. These methods, tailored to your needs, enhance your data's quality and usability.

Politecnico di Milano (POLIMI)

Evaluation of results of physical testing

Università degli Studi di Milano

This service performs the crucial step that follows the execution of physical experimentation, i.e., evaluation of the performance of the system under test. This activity requires the application of suitable performance metrics to specific test data collected during the tests and then interpreting the results in view of the features of the system and the experimental setting. Depending on the specific use case, performance metrics can involve pure computer processing, human expertise by agronomists, or a combination of both. Application of performance metrics may include the development of custom software to extract the necessary information from experimental data and/or to apply suitable processing to the information. If the performance metrics require the data to be subjected to some form of pre-processing, Service S00115 may be leveraged to prepare the data. Besides the object of this service (i.e., evaluation of results), AgrifoodTEF can, on request, support the customer along any other aspect and phase of the physical experimentation pipeline. For instance, for the design of the environment and protocol for the test, the customer can leverage services S00106 and S00107. Preparation of the test environment can be done either by the customer or by AgrifoodTEF via service S00110; in both cases, assistance in interconnecting the system under test with AgrifoodTEF’s test infrastructure is available via service S00111. Finally, data collection during the test can, if needed, be provided by AgrifoodTEF via service S00113.

Data analysis

Feasibility analysis of a selected AI or robotics use case in agriculture

This follow-up service supports clients who, based on a prior assessment or internal selection, want to explore a specific AI or robotics use case in greater depth. It focuses on understanding the conditions under which the selected use case could be applied in agriculture. Our team performs a structured feasibility analysis considering technical, organisational, and sector-specific factors. This includes evaluating compatibility with agricultural processes, identifying potential barriers to adoption, and estimating the expected added value. The goal is to provide the client with a clearer picture of what successful implementation would require and which factors might influence success or failure. The deliverable is a detailed report with insights, risks, and recommendations, supporting further internal evaluation and strategic planning.

Desk assessment

Market research

Performance evaluation of an AI model

Our service provides a thorough evaluation of AI models by utilising a wide range of benchmark datasets—from images and sensor data to remote sensing and multimodal data. This comprehensive testing assesses the model's performance across standard metrics, specifically chosen based on the model’s intended tasks. The results offer insights into the model’s effectiveness and accuracy, guiding developers in refining their AI solutions to achieve optimal performance in real-world applications.

Provision of datasets

Our service offers a comprehensive dataset tailored for training or testing your solutions in the technology and agricultural sectors. Whether you're looking to enhance existing models or build new ones, our versatile data can be adapted to fit your specific needs. For those requiring an even more robust solution, our service seamlessly integrates with our data augmentation and labelling service (S00172), ensuring that you have access to well-organized and precisely labelled data. This flexibility allows us to support a variety of sensors and use cases, utilizing both existing data records and capturing new data in your application area. Optimize your agricultural technology solutions with our data services, designed to propel your projects forward efficiently and effectively.

Provision of datasets

Testing of AI-based sensor performance

Our service thoroughly evaluates AI-based sensors by measuring key metrics like accuracy, mean squared error, and other tailored parameters. Testing is conducted in real-world environments where the sensor will be used, with reference data recorded or labelled for analysis. For certain sensors, we also utilise benchmark datasets to validate performance. We assess consistency under identical conditions, adaptability to environmental changes (e.g., light, temperature, or weather), and improvements in accuracy over time. Additionally, we measure power consumption, processing efficiency, and memory usage to ensure optimal resource utilisation. Response time between detection and action is evaluated for real-time applications, while extreme condition testing (e.g., bright light, rainy conditions, compacted soil) ensures robustness in challenging environments. We also verify data security and resilience against adversarial attacks. This comprehensive testing is essential, ensuring sensors meet high standards of precision, reliability, and efficiency. Benchmark datasets, standardised data collections used for validation, provide a reliable baseline for performance comparison. By addressing these aspects, our service ensures your sensor technology is ready for real-world deployment, delivering the performance and adaptability your application demands.

Collection of test data

Test design

Test execution

Test setup

Testing of autonomous carrier & implement integration

Evaluating an agricultural implement on an autonomous carrier platform involves more than just numerical metrics; it includes a qualitative analysis of various aspects. We assess work quality by examining performance under different agricultural conditions, including, among others, different soil moisture levels, varying crop densities, and sloped terrain. Safety evaluations consider system robustness in various scenarios. Functionality assessments measure adaptability in dynamic environments, accounting for sudden weather changes and unexpected challenges.

Collection of test data

Data analysis

Test design

Test execution

Test setup

Testing of biomass and yield estimation

Our testing service assesses the accuracy of your system's biomass estimation capabilities. You can opt to evaluate your model virtually using existing datasets, such as satellite or image data, or we can gather new field data with our field sampling equipment specifically for your needs. The main targeted crops of the biomass estimation are grassland as well as clover and alfalfa, but also grain crops are possible. As yield and biomass are highly correlated in grassland, both can be determined with this test. We provide results using metrics like Pearson correlation/R-squared and MSE/MAE to give you a clear picture of your model's performance.

Collection of test data