Research On The Image-based Transmission Method Of Settlement Pose Of Subgrade Surface For High-speed Railway

The minor deformation of subgrade of railways,tunnels,bridges and infrastructures may cause highway cracks,tunnel cracks,and even highway collapse.Therefore,grasping the subgrade settlement is crucial for the safe running of these infrastructure.But the traditional methods of subgrade settlement have the defects of low monitoring precision and maintenance difficulties,and the subgrade settlement monitoring technologies based on imaging processing can overcome the dark side of the traditional methods,it make the great range,automatic,non-contact,real-time and high-precision subgrade settlement monitoring come true by camera measurement and pose transfer measurement.The related problems in the long-term monitoring process of pose transfer measures of subgrade settlement based on image are studied in this thesis,and the research conclusions are applied to this system.Firstly,the camera imaging model is analyzed according to the basic theory of camera measurement,the pinhole imaging model and the nonlinear model are explained in detail on the basis of analyzing the transformation relationships of different coordinate systems,and the camera parameters are obtained by camera calibration.Then,the principle of the two-stage pose transfer measurement method is studied,and the principle of the measurement unit of the pose transfer measurement system is analyzed.Finally,the mathematical model of the multi-level pose transfer measurement system and its principle are analyzed.Secondly,the pose measurement system of monitoring station is designed because the deflection of the relative position of camera and target occurs in the actual operation circumstances.This system mainly designs the feature-points target which is different from traditional target.According to the feature-point target,the regulation of the camera-target relative position and pose measurement can be realized.The validity of the pose measurement system and the method is also verified by experiments.Thirdly,according to the actual environment in which the monitoring system is installed,the main factors affecting the image quality are analyzed.The primary factors are the light source and the receiving target of the system.Then the image quality evaluation method of the monitoring system is acquired by analyzing the characteristics of the light spot image.At last,the effectiveness of this method is testified through the experiments,and the optimal average gray value of the spot image is studied by screening.Eventually,the monitoring system designed in this thesis is installed on both sides of the railway track bed,and the spot image is filtered using the optimal spot image quality gray value,and a pose measurement system is used to correct the deflection between the camera and the target.In the end,the pattern of subgrade settlement during the installation experiment is obtained through analyzing the data collected by the monitoring center.The results manifest that the monitoring system can sufficiently adjust the deflection between the camera and the target,and the accuracy of the improved monitoring system is further promoted.In conclusion,the improved monitoring system comes true the non-contact,automation and high-precision measure in the true sense.