Abstract

A real machine vision set-up which detects glue spots inside plastic lids by using active thermography is presented. An infrared camera ElectroPhysics PV320, uncooled BST focal array with a bandwidth extended from 2 to 14 μm, has been used. The different parameters characterizing the defect and the active thermal technique are studied to obtain a comprehensive assessment of the phenomena and optimize them. Image processing tools such as noise filtering phase, glue thermal print extraction, and segmentation stage, are implemented to develop an automatic decision system. This process was applied on a test sample set, all the lids without any glue were detected and no misclassification occurred.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call