Research Of Techniques For Wheelset Rim Thickness Detection

Railway transportation plays an important role in our country’s transportation, inparticular recent years, the great increase in speed makes the rapid developmentof the railway transportation enterprise, the security risks brought by higher speed hasalso gained attention. Wheel is the most important part of the vehicle, and is closelyrelated to the safe operation of the vehicle, so it is very necessary to regularly detectwheel parameters. Railway departments issued relevant documents about wheelsetparameters detection, which is proposed to detect the rim thickness parameter index.In this paper, the design of online vehicle rim thickness parameter detectionmethod based on structured light imaging principle, eventually completed the solutionformula of parameters are deduced and simulated experimental verification.The main research contents in this paper are as follows:1. Put forward the structure of triangulation light detection principle applied tothe online detection device based on the rim thickness parameter, and according tothe formula, the calculation formula of the rim thickness parameters could bereceived.2. Described the key image registration algorithm in the detection method indetail. According to the image characteristics, the registration space transformationmodel was selected appropriately, and completed the realization process of derivationand algorithm of image registration algorithm in practical application.3. Designed the implement method based on the detection schemes aboutparameter was determined, including four modules: image acquisition, cameracalibration, image processing and parameter calculation. Descriptions and derivationswere carried on in detail in every module according to the key words, the formulasand the parameters.4. Several test verifications about accuracy of every module that could influencethe result of calculation, mainly included image registration module and cameracalibration module and accuracy verification module. Image registration modulemainly carries two cameras to capture image coordinates, and then solve theregistration transformation matrix of coordinate transformation, to verify the accuracyof the method. Camera calibration module mainly is validated for conversion between image points and space points accuracy. Parameter calculation module is mainly toverify the final solution accuracy dimension simulation rim thickness detectionmodel.