Blister pack inspection: Datasets & Models

Manual inspection of blister packs is slow, fatiguing, and inconsistent, and in a pharma context a single missed empty cavity or broken tablet can trigger recalls and regulatory findings. A detection model checks every cavity on every pack without the attention drift that affects human inspectors over a shift. It catches missing pills, cracked tablets, and seal defects earlier, reducing waste of foil and active ingredients and preventing defective product from reaching downstream packaging. The same approach carries over to semiconductor tray and blister-style component inspection where similar cavity-fill verification is needed.

Teams get consistent, 100% inline inspection that integrates with existing line cameras and PLC reject mechanisms rather than adding a separate manual station. Object detection scales across product formats and cavity counts by retraining on new annotated images, and RF-DETR delivers strong accuracy on small, densely packed objects like tablets in adjacent cavities. The result is fewer escaped defects, lower rework, and an auditable inspection record.

1. Upload an image: Drop any photo into the widget above to try blister pack inspection for free, right on this page.
2. Preview instant results: The model detects tablet, capsule, empty cavity, broken tablet, foil seal within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
3. Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR on your own images for production accuracy.
4. Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.