FIH Mobile used Google’s Cloud AutoML Vision and reduced their defect escape rate to 10%, while the time to inspect each component decreased dramatically to 0.3 seconds, or 1.3 seconds once fixture movement time is taken into account.
This is really amazing! And you can read the article here.
Has anyone in our community used AutoML Vision? Can you share with us the idea of the project, the issues (if any), and how you overcame them?