An intelligent vision system for detecting defects in glass products for packaging and domestic use

Abstract

Defect detection is an important task in glass manufacturing. Despite the importance of the visual inspection of glass products, many of the visual inspection processes are performed manually. The problem is that human inspection presents some drawbacks, such as being time-consuming, the high cost involved, and the lack of standardization. In this context, the development of automated processes for the inspection of glass products is important. In this paper, we propose an intelligent vision system for the automatic inspection of two types of defects in glass products: the first one is detection of a critical defect in glass cups for food packaging and domestic use, called glass sparkle or fragment of glass, and the second one is identification of a defect called deformation in plates for domestic use. To evaluate these applications, we used an apparatus consisting of a conveyor belt and a camera controlled by a PC to simulate an industrial line of production. The results indicate that the developed applications are suitable for the detection of investigated defects because for both applications, the hit rate was above 95 %.