SciTransfer
CrackSense · Project

AI-Powered Early Warning System to Prevent Fruit Cracking and Yield Loss

foodTestedTRL 5

Imagine knowing your fruit is about to crack before it actually happens, almost like a weather forecast for every single tree. By combining satellite images, drones, and ground sensors, this system spots the tiny signs of stress in the peel. It helps farmers adjust watering and care in real-time to save their harvest from spoiling.

By the numbers
50%
reduction in yield losses
50%
increase in financial gain for farmers
The business problem

What needed solving

Fruit cracking causes massive, unpredictable yield losses in citrus, pomegranate, grapes, and cherries. Current methods cannot predict when environmental conditions will trigger this peel disorder before it becomes visible.

The solution

What was built

A multi-level sensing system and prediction models that combine satellite, drone, and ground sensor data to assess cracking risk from the individual fruit level up to the regional scale.

Audience

Who needs this

Commercial citrus and cherry growersAgri-tech sensor manufacturersPrecision farming software developersAgricultural risk consultants
Business applications

Who can put this to work

Precision Agriculture
SME
Target: Agri-tech software provider

If you are a software provider dealing with imprecise crop monitoring — this project developed a sensing and digital data technology that predicts cracking risk at fruit, tree, and regional scales. This allows for the creation of a decision support system for growers.

Commercial Fruit Production
enterprise
Target: Large-scale orchard owner

If you are an orchard owner dealing with unpredictable yield losses in citrus or cherries — this project developed a real-time monitoring solution that can reduce yield losses by 1/2. This leads to 50% higher financial gains for the grower.

Agricultural Insurance
mid-size
Target: Crop insurance firm

If you are an insurer dealing with erratic crop damage claims — this project developed risk assessment models based on Earth Observation and sensor data. This provides a data-driven way to monitor agricultural production and risk levels across regions.

Frequently asked

Quick answers

What is the expected cost or price of the solution?

Based on available project data, specific pricing or cost structures for the end-user are not provided.

Can this be scaled to an industrial level?

Yes, the project specifically focuses on upscaling sensing technologies across four levels: fruit, tree, orchard, and region, utilizing EU-wide data sets and Copernicus satellite data.

How is the intellectual property or licensing handled?

Based on available project data, there are no specific details regarding the IP or licensing agreements for the developed models.

When will the solution be ready for market use?

The project period runs from 2023-01-01 to 2026-12-31, suggesting the final validated models will be available by the end of 2026.

How does this integrate with existing farm equipment?

The system integrates various data sources including Earth Observation (Sentinel), meteorological stations, and proximal sensors to provide real-time feedback.

Consortium

Who built it

The consortium is highly market-oriented with a 37% industry ratio, comprising 19 partners across 7 countries. With 7 industry partners (including 5 SMEs) and 8 academic/research entities, the project is well-balanced between theoretical modeling and practical commercial application.

How to reach the team

Contact the Volcani Centre in Israel for technical specifications on the sensing models.

Next steps

Talk to the team behind this work.

Contact us to connect with the CrackSense consortium for early access to prediction models.

More in Food & Agriculture
See all Food & Agriculture projects