Overview
Validation of fruit detections, diameter estimation, and yield estimates using artificial intelligence, in particular computer vision techniques.
The service will help the solution provider check its performance using reference datasets or experiments conducted in testing fields. The service might also include the generation of reference datasets and expected detection, diameter measures, or yield values, which can be used both for validation and training of the artificial intelligence solution based on computer vision.
More about the service
How can the service help you?
The service helps solution providers validate the accuracy of their yield estimations, fruit detection, and diameter estimation using AI and computer vision techniques, ensuring reliable performance through reference datasets and testing.
How the service will be delivered
The validation service will be conducted in designated testing fields or at your location, depending on availability and requirements. The execution of the service will take up to two weeks, allowing time for data collection and analysis. You will receive a comprehensive report detailing the validation results, including metrics on fruit detection accuracy, diameter estimation, and overall yield estimates. Additionally, metadata related to environmental conditions during testing will be provided. To facilitate this service, customers will need to supply access to relevant images or videos of fruit trees, as well as any specific criteria or parameters for the validation process. This service can be delivered year-round, although timing may be influenced by the fruiting cycle of the trees being assessed. However, you will also have the option of validating your yield estimation solution in a completely digital and simulated setting. Using the digital infrastructure of our data space, you will be able to run your yield estimation service against our datasets of images for different tree crops and get reports about the performance obtained that will help you further improve your solution.
Service customisation
The service can be customised for your specific product.
Sector
Greenhouse
Horticulture
Tree Crops
Viticulture
Test type
Data
Software or AI model
Type of service
AI model training
Collection of test data
Performance evaluation
Test design
Test execution
Test setup
Country of delivery
Country of delivery:
At user's premises
Remote
Service provider(s):
-
Universitat de Lleida (UdL)
| Website