Skip to main content
European Commission logo print header

Artificial intelligence for yield estimations at fruit orchards

Article Category

Article available in the following languages:

On-tree fruit detection system aids precision agriculture

Highly accurate fruit counting system uses artificial intelligence to help farmers optimise harvests and plan market sales.

Food and Natural Resources icon Food and Natural Resources

Farmers need to know what their expected yields are, so they can plan appropriately. But counting fruit is a time-consuming task. The EU-supported AGERPIX project has devised a more efficient way for agricultural suppliers to work: an automated on-tree fruit counting system. It lets producers assess how much fruit they will grow at accuracy levels of up to 95 %. The system developed by AGERPIX gives farmers tree-to-tree orchard assessment, determines the vigour of plants, offers a leafiness index and lets them find out exactly which zones are most productive. Using cameras, precision positioning and artificial intelligence (AI), it allows farmers to gather the true production values of their farms. There are multiple benefits to gain from the system, including orchard management, organising logistics fleets, optimisation of cold storage, shipping and HR teams, improving data processing and costs, and perhaps most importantly, an improvement in the quality of fruit. “In short, the objective is to make plantations produce more, better and with fewer resources,” says David Frances, CEO of Agerpix technologies, the project host.

Roving detector

The system goes through plantations, row by row, with an array of sensors carried on a quad bike. The data taken in through the system is then analysed and processed with AI, to identify the separate fruits on each tree. The results are then delivered to the producer and are accessible on any device. “AI is a technology very applicable to crops, it allows us to generate predictions of harvest, phenological status of the fruit, plant vigour and possible pests,” notes Frances, AGERPIX project coordinator. The system is capable of identifying the physiological state of the plant, which allows it to detect nutrient deficiencies, water stress and diseases. All data, records and production analysis are stored securely and privately in the cloud. A ‘nighttime operations’ mode lets the system work in all conditions. High-output, low-energy LED lighting means it can operate at night and not interfere with other farming tasks. The embedded GPS component provides highly accurate geo-referenced data, which separates out plots of land according to corresponding production values. This is an incredibly useful tool for farmers in seeing where the strengths of their land lie. The product is currently offered as a service, but the company plans to develop completely autonomous data collection systems in the future.

Feeding the future

“With the growing demand for food that society will have, the increase in population, the problem of insufficient water and the impossibility of new cultivation fields, it is essential to increase the productivity of plantations, and that is our fundamental objective,” explains Frances. The EU grant helped the team generate and focus their business plan, which they are now progressing to put into practice. Agerpix will continue to generate value through their information, Frances adds: “For this we are developing new ways of helping producers to make better decisions.” The company is developing the technology further, to increase the number of species and varieties the system is able to detect. “We continue to develop other types of products related to precision agriculture, solving problems raised by the producers themselves,” says Frances.


AGERPIX, crops, software, fruit, detect, precision, agriculture, AI, artificial intelligence

Discover other articles in the same domain of application