Challenges Faced

We are facing similar challenges in several ongoing projects.
All of these projects have received a satisfactory score for "ability to detect good products" during the evaluation phase, but when it comes time to put them into the field, we are faced with the same challenge: "How do we detect good products?

1. How to register good products efficiently in daily operations
2. How to efficiently perform visual checks of non-good items

is the problem.
It's not about image processing any more.
Camera and lens selection, lighting techniques, image processing algorithms...
It may be "unimportant" at the field level.