Font Size: a A A

Research On Binocular Vision Measurement Technology For High-speed Rail Contact Line

Posted on:2020-09-29Degree:MasterType:Thesis
Country:ChinaCandidate:H WangFull Text:PDF
GTID:2432330590962293Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Contact wire is a key component of high-speed rail catenary.The electric locomotive obtains electric energy through the frictional contact between the pantograph and the contact wire.The contact wire parameters?contact-wire height,stagger and wear?directly affect the stability and reliability of the power supply of the locomotive,and even affect the safety of the locomotive operation.Therefore,it is necessary to regularly check the contact wire parameters to ensure that the values of the parameters are within the scope of the technical specifications,and repair or replace the unqualified road sections in time.In this paper,the research status of catenary and contact wire detection at home and abroad,especially the application research of visual inspection technology in this field,has been comprehensively reviewed and analyzed.On this basis,the binocular measurement system is combined with the traditional gauge ruler to complete the design of the portable multi-function contact wire parameter measurement system,and the theoretical analysis and measurement experiment of the system is carried out.Firstly,according to the practical application scenarios and measurement technical requirements,the mathematical model of the contact wire binocular measurement system is established,and the functional relationship between the contact-wire height and contact-wire stagger values and the camera internal and external parameters and the relative position parameters of the two cameras is derived.The influence of the camera parameters and the relative positional relationship between the two cameras on the measurement results of the contact wire parameters is analyzed in detail.Secondly,Zhang Zhengyou's two-step method of planar template is used to calibrate the camera.The monocular camera and binocular camera calibration program are designed independently,and the calibration experiment is carried out to obtain the internal and external parameter matrix,distortion coefficient and relative position relationship of the two cameras.Based on the binocular camera calibration results,the inverse coordinate kinematics analysis of the camera coordinate system is performed to further determine the specific rotational positional relationship between the two camera coordinate systems.Thirdly,the SURF feature matching method is used to match the left and right images,and according to the mismatch problem in feature matching,the method of combining RANSAC method with Polar wire constraint is used to remove the mismatch,and the matching experiment is effective.On this basis,feature matching has higher accuracy and reliability,The measurement accuracy can be improved by using the matching point coordinate x1 and x2 with high matching precision to calculate the contact wire parameters.Fourthly,with Visual Studio 2015 as the development platform,based on the OpenCV3.2.0 function library,the program design of each module of the binocular measurement system is completed by C++language.Finally,the experimental device for the contact wire binocular measurement system was designed and manufactured independently,and the actual scene is simulated to measure the physical object of contact wire.The experimental results show that within the effective measurement range,the measurement error of the contact-wire height isą6.103mm,and the measurement error of the contact-wire stagger isą1.983mm;in the central area of the effective measurement range,the measurement error of the contact-wire height isą3.091mm,and the measurement error of the contact-wire stagger isą0.962mm.The experimental results verify the feasibility of the contact wire binocular measurement system and provide an important basis for practical instrument development.
Keywords/Search Tags:Contact wire, Binocular vision measurement, Camera calibration, Feature extraction, Feature matching
PDF Full Text Request
Related items