Automatic visual inspection system for stamped sheet metals (AVIS3M)

This paper presents a development of an automatic system for visual inspection of stamped sheet metals. Visual inspection operations in manufacturing processes can be challenging for humans. Human operators have several disadvantages compared to intelligent machines including; subjectivity, productivity, consistency, repeatability, etc. In this study we present an investigation of an automatic visual inspection system to improve some of these deficiencies. Our approach proposes an integrated software system based on machine vision theory that includes: image preprocessing, image segmentation, connected component analysis, region analysis, feature extraction, defect detection, quantification and classification and decision making. Our proposed demonstrated superior efficiency, reliability and stability compared to human subjects.