Implementation of automatic inspection system

Abstract

Inspection is the formal examination of a product manufactured by any manufacturing process. Inspection process is critical to all the industries as it ensures that a good quality product reach at the end customers. In order to ensure minimum variation in the quality of inspection of a product, industries rely on advanced equipment or gauge to measure the quality parameters of the product. The accuracy of inspection depends a lot on the method and equipments used to inspect a product. However, at AvCarb material solutions, a product called pyrolytic graphite sheet (PGS) is manufactured and at present the types of defects that occur during their manufacturing process can only be identified visually. The problem with using human senses as a tool to perform an inspection is that the accuracy and speed of the inspection vary from person to person based on their experience, their state of mind and other human factors. Automating a visual inspection system ensures minimum variation in the accuracy and speed of an inspection process. This thesis proposes the use of the automatic vision system to perform visual inspection of PGS. The document presents how some software tools can be used to identify and quantify the defects generated on PGS and gives a comparison of the accuracy of identifying a defect through the automatic vision system and accuracy achieved through human inspection.