Sistema de apoio operacional aplicado à inspeção da qualidade visual de produtos de vidro para embalagens e uso doméstico

Abstract

An important aspect in the process of glass production is the difficulty to predict all possible problems or defects before the finished product, being necessary a continuous inspection process of products. In addition, some defects beyond the issue of visual quality of the product. It is the case of glass products for food packaging, which can offer risks to the health of consumers. Despite the importance of visual inspection process of glass products, many of them are still performed manually. The problem is that manual inspection has some drawbacks such as high consumption of time, high cost and lack of standardization. In this context, the use of information technology tools applied in automatic visual inspection of glass products can provide a competitive advantage. In this paper it was proposed an operational support system employing visual pattern recognition techniques for automatic inspection of defects in glass products, which comprises two applications. The first one for detecting a specific type of defect known as spark or fragment glass in products used for food packaging and the second one to identify a defect, called deformation in household dishes. To evaluate the applications we used an apparatus consisting of a conveyor belt and a camera controlled by a PC in order to simulate an industrial line of production. The results indicate that the developed applications are suitable for detection of investigated defects, since in both applications the hit rate was above 95%.