Heuft Oceania has integrated AI-based optical character verification (OCV) into its inspection systems to check best-before dates, batch and lot numbers on packaging, including when inkjet print quality deteriorates.
The Heuft reflexx A.I. camera and image processing system uses a deep-learning model trained with images of relevant coding variations. It separates and centres individual characters before assessing letters and numbers.
The technology was integrated into the Heuft canLine II for high-speed empty can inspection and demonstrated at interpack 2026 as part of the Heuft Prime full container check.
According to Heuft, the system can verify printed information despite issues including misaligned ink dots, distorted, smudged or blurred characters and lines of text. It can operate at line speeds of 1,200 packs per minute and above.
The company said the system can reduce false rejections caused by declining coding quality, allowing printer maintenance to be postponed until scheduled line stops rather than requiring an immediate repair or production stoppage.
The Heuft Prime full container check also uses lateral profile measurement to examine the tightness of freshly filled beverage cartons.
The Heuft reflexx A.I. camera and AI-based image processing can be integrated into Heuft's modular inspection systems for applications including deep-learning OCV.
