An article in the June 28th edition of the Machine Vision Online Newsletter discusses how intricate the process of teaching a machine vision system can be. As the author points out, the system must be taught about the object under test and then programmed to make the right decisions.
Contributing AIA Editor, Winn Hardin, explores the process with Teledyne DALSA's Ben Dawson. The article highlights the effort that has to go into using a CAD drawing or HDL part description and the potentional to improve the defect-detection process through "teaching by sample."
Read the full article at Automated Imaging Association.
Teledyne DALSA's iNspect software selects multiple features to classify a feature as a scratch versus good threads.