The amount, chemical composition and distribution of nonmetallic inclusions are important factors when determining steel quality. Therefore, in recent years, a great deal of effort has gone into developing robust detection systems for nonmetallic inclusions. Various methods have been suggested, but most of them require extensive sample preparation. As a result, these methods are only suitable for analyzing micro-inclusions. For estimating the macrocleanliness, these methods are too slow because, due to the rarity of these kinds of inclusions, huge sample volumes are needed. To overcome this problem, we propose a new detection methodology that can detect inclusions and defects, like microcracks and shrinkage holes, at a range of between 5 μm and several thousands of a μm. At the same time, it is fast enough to process samples of very large steel volumes of 300 × 120 × 90 mm³ in size. Similar to the classic metallographic approach using a microscope and polished steel surfaces, the steel surface of a sample is recorded by a moving CCD sensor with a pixel size of 2.75 μm. Inclusions are detected using image processing techniques. Then, the steel surface is milled off, removing a 10 μm chip, and the recording step is repeated. Processing the whole sample in this way allows us to reconstruct the three-dimensional shape of inclusions and other defects and gives us information about the spatial distribution of these inclusions. In this paper, we describe this system in greater detail and describe the initial results of our approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.