Image
Overview
Support clients in defining metrics and testing their AI solutions in real life, in the field, or virtually. Within normal and "stressed" operating conditions, including AI-generated test cases, we assess the readiness of agriculture-related AI algorithms for classification (weeds, chromatic analysis of crops), surround understanding and perception capabilities (for safety, for navigation), yield measurement (quantity of produce, quality), and robotics (anomaly detection, etc.). Besides accuracy metrics, we have experience evaluating memory and computing power footprints, and by doing so, understanding bottlenecks and improving the energy efficiency of proposed solutions.
More about the service
Discover more about our service, including how it can benefit you, the delivery process, and the options for customisation tailored to your specific needs!
This service allows the client to assess the performance of their algorithms by testing and comparing them with benchmark solutions, including state-of-the-art models and other solutions available on the market.
The service can support the client in transitioning from a prototype version to a more stable and robust one, aiming at obtaining a market-ready solution.
The service can support the client in transitioning from a prototype version to a more stable and robust one, aiming at obtaining a market-ready solution.
The service execution lasts approximately 16 weeks, adjustable as needed.
There are no specific restrictions due to the vegetation period.The service can be executed remotely once the customer provides its algorithms and targets the optimisation needs in dedicated meetings.
The service execution would benefit from customer provision (optional) of data or should be linked to a dedicated service offered by FBK for data provision (S00353).
There are no specific restrictions due to the vegetation period.The service can be executed remotely once the customer provides its algorithms and targets the optimisation needs in dedicated meetings.
The service execution would benefit from customer provision (optional) of data or should be linked to a dedicated service offered by FBK for data provision (S00353).
Customisation for this service is possible given the wide spectrum of algorithms that can be used for many different purposes (i.e., anomaly detection, navigation software, fruit recognition). A limitation is that the algorithms must be executable on conventional hardware, either in the cloud or locally, without the need for dedicated hardware.
Location
Remote
Type of Sector
Arable farming
Greenhouse
Horticulture
Livestock farming
Tree Crops
Type of service
Performance evaluation
Test design
Test execution
Test setup
Accepted type of products
Design / Documentation
Software or AI model
