Corrosion, the natural and irreversible process that converts refined metal into a more chemically stable form such as oxide, hydroxide, carbonate. Ultimately the purity of metal goes down and it will cause failure in Welding,bindings in an industry. Accidents due to mechanical loss of metallic bridges, cars, aircraft, etc may cause damage to not only metal but the human digestive tract, eyes, skin,respiratory tract. The manual checking on these metals for corrosion detection is time-consuming. Hence this study builds the Design and Development of a system for Corrosion detection using the image segmentation algorithm U-Net. Dataset creation is first and foremost, a crucial step whenever we go for applying machine learning for any new task. During this experiment, the dataset was generated by capturing images from surrounding various devices and combining datasets from various online sources consisting of rusted metal. This approach obtains a good predictive result with an intersection over union score of 0.6160, and a Jaccard loss of 1.1356 for rust detection. In this paper, we proposed the automated technique by which we can be able to detect rust on metal.
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.