Poorly positioned, incorrect or bent components on an assembled circuit board are difficult to detect with conventional image processing. Using Deep Learning, classification and error detection is much more reliable, even with a small amount of training data.

Tasks

• Checking the correct assembly of printed circuit boards
• Positioning
• Completeness check
• Detection of faulty components