| With the continuous development of China’s railway system, the speed of the train isaccelerating. The wheel is one of the most important composition of the train, its working state hasimportant effect the safe driving to the train. So this situation puts forward higher requirements forthe testing method of the wheel and its precision. With the development of computer technologyand CCD sensor technology, the technology of wheel online detection based on structure light hasbecome one of the focus of the research on the field of wheel detection. In wheel online detection,calibration technology is one of the key technologies. It will influence the results.The main content of the research in this article:(1) According to the principle of wheel online detection, radial distortion imaging model wasestablished. Based on this model, using coplanar feature points, Tsai non-coplanar calibrationalgorithm was presented. According to the first-order radial distortion formula, Householdertransformation method and Levenberg-Marquardt optimization algorithm was used to calibrate thecamera’s internal and external parameters.(2)Characteristic parameters of the devices in wheel online detection were designed, includingCCD camera and the laser source. And according to the site environment, the position parametersabout the laser source and the camera was designed.(3)The calibration method of camera and laser sources was designed, including the calibrationboard, its locations and installation base.(4)The camera calibration software was designed. It integrated all algorithms, including imageacquisition, image display, image processing and Tsai non-coplanar calibration algorithm.(5)According to the on-site installation location situation, the simulation experiment of thecamera calibration was designed. And this paper completed the camera calibration and erroranalysis. |
PDF Full Text Request