OCR code reading on flexible packaging
Accurately read codes on bags and pouches with complex designs
Graphical programming environment for deep learning-based industrial image analysis
Consumer products are increasingly shipped and sold in flexible packaging, including bags, gusseted pouches, flat pouches, and wraps. Frozen vegetables have long been sold this way, but cosmetics, medications, powdered products, cleaning products, meats, coffee, and a wide range of other products now come in printed flexible packaging as technology has improved and prices have dropped.
Each package needs to include printed date, lot, and other specific tracking information. If the information is misprinted, or misread, supply chain problems can ensue, leading to misdelivered products, or those that miss their sell-by date, leading to waste.
Reading codes on flexible packaging is extremely challenging for conventional optical character recognition (OCR) technology. The characters can easily be printed skewed. Even when printed properly, shifts in the material can lead to variations in lighting, and distortion of the characters when being read. This is particularly true as packaging becomes more customized, and new printing techniques allow for more of the surface to be covered by product design. Conventional OCR software must be reprogrammed for each new package design.
The AI–based OCR tool reads printed codes on challenging flexible substrates, even with confusing printed backgrounds. The OCR tool comes with a pretrained font library, making it easy to set and deploy. The OCR tool trains on a small set of images of skewed, angled, unevenly lit, and deformed text and then finds and reads such text on flexible packaging, no matter how it is shaped or creased.
Products in flexible packaging reliably and promptly reach their designated customers.