Robotics, vision and AI combine for dairy processes
Automation solutions provider Eberle Automatische Systeme has developed a breakthrough in quality control with a focus on the cheese-ripening process, combining AI with machine vision software from MVTec.
The food industry is experiencing a transformative shift in quality control, due in part to advances in artificial intelligence (AI). When combined with rule-based machine vision, AI is enabling automation of processes that were previously impossible, unlocking new levels of productivity and quality assurance.
Cheese consumption is booming globally, and producers are facing increasing challenges as they scale production. Labour shortages, particularly in Europe, are pushing dairies to adopt automation to increase efficiency. Meanwhile, sustainability is becoming a central concern, with an increased focus on reducing waste and conserving resources. Additionally, consumers are demanding higher-quality products with more variety, further intensifying pressure on producers.
As Eberle’s Machine Vision Engineer, Dorian Kopfle, explains: “The cheese-ripening process, which can last up to 14 months, requires constant monitoring to avoid mold and ensure quality. Manually inspecting thousands of cheese wheels is virtually impossible, which is why Gebr. Baldauf, a traditional dairy, turned to us for an automated solution.”
Automation with machine vision and AI
Gebr. Baldauf, located in the Allgau region, commissioned Eberle to solve these challenges. The result is a fully automated monitoring system, that combines a mobile care robot, cameras, and onboard image processing.
The process begins with the inspection of cheese wheels for defects, such as mold spots or blemishes. A 4K camera captures high-resolution images, which are analysed using advanced machine-vision algorithms from MVTec HALCON. The software uses deep-learning methods to detect anomalies earlier, minimising process deviations and waste. The data is stored and made available via a web interface, enabling remote monitoring and control. Simultaneously, the mobile care robot performs its task of treating the cheese wheels, ensuring proper rind formation and removal of unwanted smear layers.
This system not only increases efficiency by reducing manual inspection but also improves the consistency and quality of the final product.
Key outcomes and business impact
The deployment of this automated system has provided several key benefits for Gebr. Baldauf, including:
- Increased efficiency: The mobile care robot operates autonomously, reducing manual labour while ensuring that each cheese wheel is inspected and treated thoroughly.
- Waste reduction: Early detection of mold or defects allows for timely intervention, preventing rejected cheese and minimising waste.
- Improved quality control: The system ensures more consistent and less subjective inspection results by replacing manual methods with AI. As a result, the process achieves a 100% inspection rate, applying the same inspection criteria throughout.
- Full traceability: The integration of industrial image processing ensures complete product traceability. All inspection results are stored digitally for easy access, enabling better decision-making and long-term process optimisation.
Overcoming technical challenges with AI
A significant challenge in developing this system was the natural variability of cheese. Every wheel looks different and undergoes significant changes during the ripening process, which makes rule-based machine vision methods less effective. To overcome this, Eberle utilized AI and deep learning to create a system that could adapt to the unique characteristics of each cheese wheel.
The MVTec HALCON software was instrumental in this process. By training a deep-learning network with a large dataset of cheese images, the system is able to reliably detect defects such as cracks, mold, and discoloration, while ignoring the natural variations inherent to the process. This technology ensures that even subtle anomalies are spotted, allowing for earlier intervention and better quality control.
Eberle’s goal was not only to automate the inspection process, but to fully integrate AI into the cheese-ripening workflow. Currently, the system is capable of performing real-time inspections and autonomous care, with minimal human involvement. However, the company is working on refining the system further to handle all types of cheese and stages of ripening, with the long-term goal of creating a fully automated, AI-driven system that requires no human input.
The system also provides a solid foundation for future digitalisation efforts, with the potential for integration into larger digital platforms, such as ERP systems and the cloud, to further optimise the production process.
Scaling and further digitalisation
Building on the success of this project, Eberle is now focused on scaling the solution to meet the needs of the entire cheese industry. The company plans to standardise the system and integrate it into both mobile and stationary care robots for cheese production worldwide.
Furthermore, the system’s AI capabilities are continually evolving. Eberle aims to refine the deep-learning models to handle different cheese types and ripening stages, enabling fully automated classification and inspection. This will allow producers to further reduce human involvement while maintaining the highest standards of quality.
As Christoph Muxel of Eberle summarises, “Our machine vision-based solution demonstrates how automation can sustainably improve quality, efficiency, and competitiveness in the food industry. This project is just the beginning, and we’re excited to take these innovations to a global scale.”
