cLynx

Artificial intelligence in industrial applications is often accompanied by data-driven process optimization. Especially in manufacturing processes, quality control can lead to bottlenecks causing costs and delivery delays.

With the cLynx family, SAUERESSIG has established a complete closed-loop system from incoming customer data and final reproduction to the production of gravure cylinders.

Digital

Our intelligent quality control software identifies potential deviations during the entire prepress stage. Maximum transparency, integration in planning tools and the highest data security requirements.

Accurate

The high degree of automation leads to a minimization of manual interactions. Traceable documentation secures the process by providing a certificate for each cylinder.

Innovative

The highest quality demands - even faster. The deep neural network generates knowledge from experience and is constantly learning. A real breakthrough in digital print image inspection.

cLynx Family

cLynx.prepress

The challenge along the prepress stage is to realize even the most extraordinary designs on a mass scale. Process-specific, optimal color separations and realistic proofs are created. The intelligent software cLynx.prepress automatically compares the customer or graphic data with the reproduction file for perfect print results.

cLynx.print

The increasing demands on design and quality are offset by a decreasing tolerance for errors. cLynx.print performs a digital comparison between reproduction data and print results. Due to an extremely fine analysis it differentiates precisely between material defects, soilings and actual production faults. This guarantees an extremely high testing quality regardless of human factors.

cLynx.cylinder

When it comes to the highest material quality, our computer vision and AI-based quality control system cLynx.cylinder looks very precisely: cameras scan the surface of our rollers and automatically detect deviations by comparing it to the expected outcome. By generating knowledge from experience with the captured images, the system learns a constantly better classification. Our deep neural network achieved fully automated classification accuracy rate of nearly 99%.