Quality inspection of engraved image using shape-based matching approach

Abstract

The role of machine vision system as a vital component for quality control mainly in manufacturing process cannot be denied. The system is developed to overcome the discrepancy from human vision and illumination changes. This paper proposes shape-based vision algorithm, a hierarchical template-matching approach that implemented in flexible manufacturing system to verify the quality of engraved image. Color and gray scale charged couple device (CCD) cameras are used to acquire engraved image for different kind of environment. The engraved image is preprocessed using image processing technique. Region of interest (ROI) is then selected and digitized into gray level to extract the contour of the object using segmentation technique. The extracted contour is used as template for object recognition during matching process. Several objects are engraved on the acrylic souvenir bases with different color background to test the algorithm. This experiment result shows that the algorithm works better with detection rate of 100% and matching accuracy of more than 98%. The approach can be applied in packaging, pharmacy, education, medical or any other areas which apply shape in their application.