Tablet and pill inspection
Automatically detect damaged pills and foreign matter on a packed conveyor
Graphical programming environment for deep learning-based industrial image analysis
Solid oral unit doses of medication often come in the form of tablets, also called pills in layman’s terms. Tablets are manufactured in a tablet press and may be coated after they emerge from the press.
Tablets are the most common form of solid oral unit dosage of medication. Before they go into their bottles, blister packs, or other packaging, they undergo visual detect detection to confirm that various features meet specifications, such as dimensions, surface texture, shape, color, freedom from surface defects, and molded or printed labeling. If any of those are incorrect, or if there is foreign material that might potentially contaminate the container, the damaged tablet or contaminating object must be removed.
Allowing foreign matter, damaged pills, or even incorrect pills into a container can lead to a recall, and a loss of sales and reputation.
Tablets are fed onto a conveyor. Both sides of each tablet must be inspected. Because of its visual complexity, tablet inspection is often manual, although human beings are quite poor at spotting these defects. There are a large number of possible defects as well as types of foreign material. Reflective coatings may lead to the product looking damaged when it is not. As a result, it is impossible to effectively use conventional machine vision for inspection, the reason manual inspection has survived.
Cognex AI-powered technology solves the problem of inspecting large numbers of tablets with high accuracy. It trains on images of acceptable tablets taken from various angles. The defect detection tool then detects any tablets with anomalies, even when not included in the original training set, while passing all acceptable tablets on to primary packaging.
Tablets reach end consumers undamaged, enhancing brand reputation and minimizing recall risk.