Print Email Facebook Twitter Semi-supervised rail defect detection from imbalanced image data Title Semi-supervised rail defect detection from imbalanced image data Author Hajizadeh, S. (TU Delft Railway Engineering) Nunez, Alfredo (TU Delft Railway Engineering) Tax, D.M.J. (TU Delft Pattern Recognition and Bioinformatics) Contributor Acarman, Tankut (editor) Date 2016 Abstract Rail defect detection by video cameras has recently gained much attention in bothacademia and industry. Rail image data has two properties. It is highly imbalanced towards the non-defective class and it has a large number of unlabeled data samples available for semisupervised learning techniques. In this paper we investigate if positive defective candidates selected from the unlabeled data can help improve the balance between the two classes and gain performance on detecting a specic type of defects called Squats. We compare data sampling techniques as well and conclude that the semi-supervised techniques are a reasonable alternative for improving performance on applications such as rail track Squat detection from image data. Subject imbalance datasemi-supervised learningrail image datarail defect detection To reference this document use: http://resolver.tudelft.nl/uuid:fe47bca8-1a70-47c7-83a0-b906ca7722fc Source Proceedings of the 14th IFAC Symposium on Control in Transportation Systems, CTS 2016, Istanbul, Turkey Event 14th IFAC Symposium on Control in Transportation Systems, 2016-05-18 → 2016-05-20, ITU Faculty of Architecture, Istanbul, Turkey Part of collection Institutional Repository Document type conference paper Rights © 2016 S. Hajizadeh, Alfredo Nunez, D.M.J. Tax Files PDF TurkeyCTS2016_S_Hajizadeh_FINAL.pdf 1.44 MB Close viewer /islandora/object/uuid:fe47bca8-1a70-47c7-83a0-b906ca7722fc/datastream/OBJ/view