| With the high-speed development of railway industry,seamless track has been widely used because of its advantages of low maintenance cost,low train vibration and long service life.External temperature changes,train rolling,starting and braking will cause internal stress on the track.When the internal stress is greater than the fastener pressure the rails will deform and produce longitudinal displacement.Due to the seamless line almost no rail joints,the rails can not be free to release internal stress,easy to appear swelling rail,broken rail,or even rollover and other major safety accidents.Therefore,it is of great significance to design a rail longitudinal displacement detection system for the safe operation of trains.First of all,this thesis designs the overall scheme of track shape variable detection system based on image processing,and expounds the measuring principle of the system.The image acquisition scheme and related mechanical structure are designed.The industrial camera and lens parameters were calculated according to the measured distance and field of view.After the test,two Hikvision 500 W pixel color industrial cameras were selected with focal length lenses of 50 mm and 75 mm,and the secondary development of the camera was completed to complete the image collection work.Secondly,a camera deflection error compensation model is proposed for the installation error of portable image acquisition equipment.This thesis focuses on the whole process of obtaining longitudinal shape variables of railway track after image processing,which mainly includes pre-processing,improved Canny edge detection,shape screening,color extraction,size transformation and other operations.After comparing the advantages and disadvantages of each database,design the database content in My SQL database management system.The system recognizes and analyzes the TWO-DIMENSIONAL code at the measurement point and then accesses the corresponding database through the network for data management.After that,in order to facilitate the operation of the system,the human-computer interaction interface of the detection system was designed in the MFC framework of Visual Studio software.Alarm the surveyor when the measured data is greater than the safe value.Using C++ programming language and Open CV image processing open source library,various functions of rail longitudinal displacement detection system are realized,and an executable program can be run under Windows system is generated.Finally,a simulated experimental platform was built and tested on the experimental platform and the real railway.Experimental results show that this method has high accuracy and reliability,and can meet the engineering needs.The measurement depth of the detection system can reach 11 m,the displacement detection range is ±35mm,and the measurement accuracy can be kept within 1mm.At present,the system has passed the test of relevant departments and signed relevant contracts. |
