Intelligent corrosion detection and rating based on faster region-based convolutional neural network

Surface of mechanical equipment often suffer from severe corrosion damage because it exposes to corrosive environments all year round. Various non-destructive monitoring methods were developed to detect the location and extent of corrosion so as to arrange maintenance. In this paper, a novel intelligent corrosion detection method is proposed by using faster region-based convolutional neural network and a new rating index is defined to evaluate surface condition of specimens, which might realize corresponding surface ratings. Finally, an experimental study on metals plates composed of several base metals and coatings commonly used in engineering application is implement to verify the effectiveness of the proposed method. The experimental results show that the proposed method can accurately detect corrosion damage on the specimens and give the correct appearance rating.