EyeC GmbH, a leading global provider of automated artwork and print image inspection systems, announces the official launch of its AI-based defect classification. Effective immediately, all inline inspection systems in the EyeC ProofRunner Web line will be equipped with the world’s first pre-trained AI. This groundbreaking technology will be showcased at Labelexpo Europe 2025 at booth 5B11 in hall 5.
EyeC’s new AI solution marks a major step forward – from simple defect detection to content based classification and analysis. The pre-trained system identifies relevant quality deviations with precision and automatically categorises them into four main defect classes: material, dust, spots, and streaks – plus the category “other”. Additional classes will be added in regular cycles. With the highest assignment accuracy, the EyeC ProofRunner Web detects all defects, even hidden material flaws within text fields. A key advantage of this new technology lies in its ability to distinguish between critical defects that require immediate corrective action, such as ink smears or material issues, and non-critical deviations like dust particles or substrate inclusions. While the latter are documented, they do not trigger unnecessary production stops.
The result: significantly fewer false positives combined with faster processing speeds.
Sustainability in focus
The AI-powered defect classification is particularly relevant for companies that must comply with regulations and initiatives aimed at reducing packaging waste and promoting a circular economy, such as the EU Packaging and Packaging Waste Regulation (PPWR). The advanced image processing capabilities consider the varying optical properties of sustainable materials, enabling the system to reliably distinguish between acceptable surface irregularities and true defects when inspecting recycled substrates.
The training methodology is based on extensive datasets comprising hundreds of thousands of real print job images. By analysing this vast image library, EyeC’s artificial intelligence learns to differentiate between flaws that require attention and those that have no impact on the final product’s quality. The pre-trained system eliminates the need for complex customer specific training and is ready for immediate use.
“With our AI-based defect classification, we’re setting a new industry standard,” says Nico Hagemann, Director of Product Management at EyeC GmbH. “While conventional systems only detect defects, our AI understands them like an experienced expert and distinguishes critical issues from irrelevant deviations. This empowers our customers to carry out precise inline inspections at full production speed while saving both time and resources.”
The AI-based defect classification for the EyeC ProofRunner Web will be available as part of its base configuration starting with Labelexpo Europe 2025 and will be included as a standard feature with every new system. Live demonstrations will be held at EyeC booth 5B11 in hall 5. The rollout to additional systems, including the EyeC ProofRunner Sheetfed, will follow in close cooperation with OEM partners. For more information, visit www.EyeC.com.