Code Reading on Large Pallets
Reliably read codes of varying size in large fields of view
Graphical programming environment for deep learning-based industrial image analysis
Crates and pallets holding large amounts of product in warehouses, yards, and other large-scale storage areas are identified by text codes for tracking and shipping. The codes are large, so they can be viewed at a distance. They are also prone to printing errors, being partly obscured, underlit or flared, and having other reading problems under storage facility or yard conditions.
Errors in reading can lead to misdeliveries and misplacement of goods. Given the size and weight of these pallets, rearranging them takes time and effort, and can delay shipping.
Up to now, such text codes have been largely read and recorded manually. While it would seem that conventional OCR technology would be ideal for this application, the large, required field of view, the variation in viewing distances leading to variation in visible text size, and the unpredictable locations of codes make it difficult to program conventional OCR software to handle all possibilities.
Cognex Deep Learning is ideal for reading codes with unpredictable sizes resulting from varying distances. The Deep Learning OCR tool comes with a pretrained font library, making it easy to set up and deploy. The OCR tool trains on a small set of images of variously sized codes, and then finds and reads such codes in the facility or yard, no matter what the distance, under conditions of variable lighting. The eliminates any need to move or direct the camera, improving efficiency.
Heavy pallets are arranged most efficiently in distribution areas and reliably delivered to their end consumers.