Using pattern recognition algorithms for printing inspection
John Stamos, 10-27-2011
In today’s more stringent regulatory climate, accuracy in printing inspection is more critical than ever. Traditional printing inspection strategies such as human “spot inspections” of random batches and using cameras to acquire and a single personal computer to process images are no longer acceptable.
One company I work met printing inspection requirements by switching to a Cognex In-Sight® vision system because it incorporates both camera and image processing in a single enclosure. The switch required only a short and simple validation procedure to ensure the new vision system provides the same functionality as the old one. The In-Sight vision system also is much easier to program because its PatMax® pattern matching algorithm easily finds the printed image with a single command. The system inspects over 400,000 tablets per hour and provides unique single tablet rejection capabilities.
The PatMax pattern recognition algorithm looks for the printed image regardless of its location in the image acquired by the camera. This eliminates the need for precision positioning of the camera or the object being inspected. PatMax identifies and isolates the key individual features within an object image and measures characteristics such as shape, dimensions, angle, arcs, and shading. It then correlates the spatial relationships between the key features of the trained image to the run-time image, encompassing both distance and relative angle. By analyzing the geometric information from both the features and spatial relationships, PatMax is able to precisely and repeatedly determine the object's position without regard to the object's angle, size, or appearance.