Syringe Plunger Stopper Inspection
Ensure stoppers are properly placed and defect free
The plunger, or piston, is the component of the prefilled syringe that, when depressed, pushes the vaccine-containing liquid out through the needle into the patient. The stopper at the end of the plunger must move smoothly in the barrel while preventing any leakage or possible contamination.
Before syringes are assembled, manufacturers must ensure that the plunger stopper is present, the correct shape, properly connected to the plunger rod, and free from particulates that can cause contamination.
Plunger stoppers have ribs, which can be torn during insertion, or have liquid between them, indicating a possible escape of the vaccine fluid, in addition to creases and other defects. Plunger stoppers are typically black, so defects can be hard to detect. Attempts to compensate by raising the contrast can introduce glare, overexposure, and other lighting challenges. The complexity is further compounded by the different colored syringe components, liquid vaccine in the glass or polymer barrel, and the location and size of the unpredictable defects.
The wide range of possible small defects around the circumference of the stopper, combined with the possible visual confusion of variations in the barrel and colored components in the syringe, make defect detection in syringe plunger stopper inspection a good candidate for deep learning. The defect detection tool is trained on a wide range of stoppers to learn to detect a variety of defects, including torn ribs, liquid between the ribs, and creases from the stopper’s insertion into the syringe barrel.
Cognex Deep Learning is equipped with High Dynamic Range Plus (HDR+) technology that provides more uniform illumination along with greater depth-of-field, making lower contrast defects on the plunger stopper more distinct without requiring complex and expensive lighting. HDR+ differs from standard HDR as it can be done with a single acquisition at high-speed on moving parts, whereas standard HDR would need to be stationary and capture multiple images to obtain the same results.