Revolutionizing Cheese Production with Artificial Intelligence and Machine Vision

[Cengiz Özemli]

## Revolutionizing Cheese Production with Artificial Intelligence and Machine Vision

The combination of industrial machine vision and robotics enables cheese producers to continuously inspect thousands of cheese wheels throughout the ripening process.

The food sector is undergoing a major transformation in quality control with the advancement of artificial intelligence (AI). AI, combined with rule-based machine vision, makes it possible to automate previously impossible processes, setting new standards in productivity and quality assurance. One of the significant innovations in this field has been introduced by Eberle Automatische Systeme, an automation leader that develops solutions for the cheese ripening process.

### The Challenge: Increasing Demand, Labor Shortages, and Sustainability

As global cheese consumption rapidly increases, producers face various challenges as they scale up capacity. Labor shortages, especially in Europe, are driving dairies to use automation to increase efficiency. At the same time, sustainability is becoming a critical issue, focusing on reducing waste and conserving resources. Consumers' expectations for higher quality and variety also increase pressure on producers.

As Dorian Köpfle, Machine Vision Engineer at Eberle, states: “The ripening process can last up to 14 months and requires continuous monitoring to prevent mold formation. Manually inspecting thousands of cheese wheels is almost impossible, which is why Gebr. Baldauf GmbH & Co. KG, a traditional dairy, requested an automated solution from us.”

### The Solution: Automation with Machine Vision and Artificial Intelligence

Gebr. Baldauf in the Allgäu region of Germany commissioned Eberle to solve these challenges. The resulting solution was a fully automated monitoring system combining a mobile maintenance robot, cameras, and integrated image processing technology.

Cheese wheels are inspected for defects such as mold spots or imperfections. High-resolution images are captured with a 4K camera and analyzed with MVTec HALCON's advanced machine vision algorithms. Thanks to deep learning techniques, anomalies are detected early, minimizing process deviations and waste. Data is made accessible via a web interface for remote monitoring and control.

At the same time, the mobile maintenance robot ensures proper rind formation of the cheeses and cleans unwanted residues. This system reduces manual inspection while increasing process efficiency and product quality.

### Key Results and Business Impact

Significant advantages offered by the automated system for Gebr. Baldauf:

• Increased efficiency: The mobile maintenance robot operates autonomously, reducing manual labor and ensuring meticulous inspection of each cheese wheel.
• Waste reduction: Early detection of mold or defects allows for timely intervention, minimizing rejected products and waste.
• Enhanced quality control: The subjectivity of paper-based or human visual inspections is reduced, with a 100% inspection rate and standardized criteria applied.
• Full traceability: All inspection results are digitally recorded with industrial image processing, enabling better decision-making and process improvement.

### Overcoming Technical Challenges with Artificial Intelligence

The natural variability of cheeses posed a significant challenge in developing the system. Each wheel looks different and undergoes significant changes during ripening, which complicates rule-based machine vision techniques. Therefore, Eberle developed a system using artificial intelligence and deep learning that adapts to the unique characteristics of each cheese wheel.

MVTec HALCON software, trained with extensive cheese image data, reliably identifies defects such as cracks, mold, and discoloration, while disregarding natural variations. This technology allows for early detection of even subtle anomalies and ensures quality control.

### Future Vision for Full Automation

Eberle's goal is not only to automate the inspection process but to fully integrate AI into the cheese ripening workflow. The current system can perform real-time inspection and autonomous maintenance, but development efforts are ongoing to cover more cheese types and ripening stages. In the long term, the aim is to establish a fully AI-powered system that requires no human intervention.

The system integrates with ERP and cloud-based digital platforms, creating a solid digitalization foundation for production optimization.

### Future Plans: Scaling and Digitalization

Following the success of this project, Eberle aims to expand the solution throughout the entire cheese industry. The system will be standardized and used in mobile and stationary maintenance robots globally. Furthermore, AI models will continue to be developed to recognize different cheese types and ripening stages. This will further reduce human intervention while maintaining quality standards.

In the words of Christoph Muxel from Eberle: "Our machine vision-based solution demonstrates how automation can sustainably improve quality, efficiency, and competitiveness in the food sector. This project is just the beginning, and we look forward to expanding these innovations globally.