Abstract
he trouble of freight car detection system (TFDS) is a popular application in Chinese railway today. In this paper, the discrete-point sampling model is further developed to locate potential fault regions in the photos taken by the TFDS. The discrete-point sampling model not only contains the image’s region boundary information and region information, but also reflects the transition from region to boundary. The most salient component’s contours in samples are drawn by hand and recorded as data templates used for matching in test images. Experimental results show that by components’ classification, the method based on this model can classify different types of freight cars’ parts universally and locate the potential fault regions more accurately and quickly than regional gray matching or edge matching. The results of anti-noise testing in laboratory and more than two years daily operation at several inspecting stations show that our method has a strong ability to survive with nonlinear deformations, and has a good extensibility to be used with different parts, which meet application demands for the full-automatic inspection system.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call