The AgroICT, Precision Agriculture and Livestock Farming group of the University of Lleida and Agrotecnio has developed FruitMeasureApp, a mobile application that uses computer vision techniques to accurately estimate fruit size directly on the tree.
The system is based on the use of a conventional smartphone equipped with a support that incorporates a 3D-printable circular element. Users simply need to bring this element close to the fruit and take a photograph. From the image, the application, developed with artificial intelligence and neural network techniques, automatically estimates the fruit’s dimensions with high precision.
This methodology allows for objective and reproducible measurements, overcoming the limitations of the traditional calliper method, which depends on the operator’s skill and can introduce measurement errors.
“Unlike other solutions available on the market, it is not based on high-cost sensors or services oriented towards large farms,” explained Eduard Gregorio, researcher in the group.
FruitMeasureApp has been developed in collaboration with the Institut de Recerca i Tecnologia Agroalimentàries (IRTA) and has been funded through the Strategic Plan of the CAP 2023-2027 by the Department of Climate Action, Food and Rural Agenda of the Government of Catalonia.
Currently, the application is in the demonstration phase, with ongoing trials in apple orchards. Preliminary results are promising, and the app is expected to be available free of charge on Google Play and the App Store in the near future.
The Escola Tècnica Superior d’Enginyeria Agroalimentària i Forestal i de Veterinària (ETSAFiV) of the Universitat de Lleida will host a demonstration as part of a technical workshop on 10 September.
More information:
· PATT Workshop: FruitMeasureApp. Mobile application based on AI to measure fruits in the field
Text: Premsa UdL / IRTA / Agrotecnio Communications