After the barrel of a syringe is cut away from the tubing, a shoulder, flange, or finger grip is formed by heating the end, rotating the barrel, and using forming tools on the soft glass. A flange can be fully round or be cut into any of a number of shapes, with small variations depending on the manufacturer and model.
The cutting, forming, and annealing processes can cause cracks, chips, or spalling, all of which can be dangerous to the user, since the flange is where the fingers are placed for injection. While scratches are typically classified as minor defects, it is important to detect even the slightest crack in the flange, since it can be a sign of a structural problem. Since these defects can be both subtle and vary in shape, size, and position, traditional machine vision can struggle to perform this step-in syringe automated inspection.
SolutionBank Deep Learning is ideal for syringe flange inspection because it can be trained on a range of flanges at various angles and in spite of the variability, transparency, and geometric complexity. It distinguishes between an acceptable flange and an unacceptable one, even for extremely subtle flaws that would be missed even by human inspection.
The SolutionBank Deep Learning defect detection tool easily adapts to slight shape variations resulting from changes in supplier that would require significant reprogramming for conventional machine vision, minimizing erroneous rejections of good flanges.
For more details, business please contact:
SOLUTIONBANK CORPORATION
Headquarter: 9 Phan Ke Binh Street, Da Kao Ward, District 1, Ho Chi Minh City, Vietnam
Hotline: +8496 113 0541
Email: sales@solutionbank.asia
