Printing defect identification in pharmaceutical blisters using image processing

Abstract

Objective: Our aim is to detect printing defects in pharmaceutical tablets from the manufacturing line using image processing techniques. Methods: The printed labels contain the details of the chemical composition, date of manufacture, date of expiry, manufacturing location, etc., images of the labels are obtained and processed using image processing algorithms to detect any defects on the labels before dispatch. Results: The printing defects on the labels such as missing letters, words, lines, and disorientation of alignments. Conclusion: Euclidean distance method was used for comparison that yielded 95% accuracy in removing tablets with printing defects.