AI tool to boost transparency in olive oil industry
Researchers at the University of Cordoba in Spain have developed an app capable of identifying olive varieties using photos of olive pits.
This knowledge can help optimise cultivation practices and ensures the production of high-quality olives. The AI tool can also help olive oil producers ensure the authenticity and quality of their products. By verifying the olive variety used, they can provide accurate information to consumers and maintain their brand reputation.The tool's development was possible thanks to the cataloguing and documentation work of five germplasm banks in different countries and to advances in Artificial Intelligence systems.
The initiative, which is part of the GEN4OLIVE European project to improve olive trees, involved the participation of olive germplasm banks from Morocco, Greece, Italy, and Turkey to gather more than 150,000 photos of 133 olive varieties from the Mediterranean basin. The Computer Science Department at Rome's Sapienza University was in charge of collecting the information and creating the algorithm for this tool, which proposes a new approach to identify varieties and automates the traditional morphological classification process.
Researchers Hristofor Miho and Concepción Muñoz Díez stressed the accuracy demonstrated by the model, with around 90% efficiency. "It's a system of learning through trial and error, based on 'Machine Learning,' in which we train the machine to learn through its own failures," they said. The researchers explained that the more images forming the database, the more effective the system will be. The entities participating in the project have agreed on very strict protocols to unify their working methodologies and generate images allowing for optimization of the algorithm.
The result is an Artificial Intelligence tool that has been shown to be capable of detecting morphological details that escape even the human eye. After processing the data, it returns a list of the possible varieties with different degrees of compatibility with the photographed sample. This Machine Learning system will be the basis of an application that will allow growers and nurseries to easily and quickly identify the variety of olive tree they are working with. According to Ucolivo, by making it available to the entire sector as a public and free tool, it will also serve to advance general knowledge of all existing olive varieties.
Image: Getty/piola666
