Huang I.-FChen P.-H.PO-HAN CHEN2021-08-052021-08-052020https://www.scopus.com/inward/record.uri?eid=2-s2.0-85102577932&doi=10.1109%2fICACEH51803.2020.9366258&partnerID=40&md5=c19e79fb8d904eb42e67dcbcb30b82c7https://scholars.lib.ntu.edu.tw/handle/123456789/576061Nowadays, bridges are significant infrastructure in most countries, and it is crucial to come up with an effective corrosion detection method for steel bridge inspection. A crucial issue on rust recognition is to distinguish real rust corrosion spots and areas. A fully convolutional neural network, namely U-Net, is explored to develop an image semantic segmentation model, which provides a wide range of rust image recognition. ? 2020 IEEE.Convolution; Hydraulics; Image recognition; Image segmentation; Network architecture; Semantics; Steel bridges; Steel corrosion; Convolutional networks; Corrosion detection; Corrosion spots; Image semantics; Rust defects; Steel bridge coatings; Steel bridge inspections; Convolutional neural networks[SDGs]SDG9conference paper10.1109/ICACEH51803.2020.93662582-s2.0-85102577932