Research On Pantograph Slide Thickness Detection Method Based On Machine Vision

Failure and accident and instability of pantograph and catenary system affect the speedup and safe operation of electrified railway,it is one of the important factors for ensuring normal operation of electrified railway.In recent years,China’s electrified railway construction has been developing towards the direction of high speed,high density and high load,which puts forward higher requirements for the operation safety of locomotive pantograph and catenary system.Pantograph slide contact catenary during train operation and cause wear.When the wear of pantograph slide exceeds the limit value,affecting the safe operation of electrified railway.However,at present,in the daily maintenance of railway system,mainly through manual measurement and visual inspection,there are low efficiency,low accuracy and easy to produce human error.To solve the above problems,this thesis proposes a method to detect the thickness of pantograph slide based on machine vision,analyze and study the thickness detection method of electric locomotive pantograph slide.The project chooses to collect pantograph slide images in the laboratory environment.The main research content is the image processing method of pantograph slide digital picture and the calculation method of pantograph slide thickness.The specific research contents are as follows:(1)In the scheme,collecting the pantograph slide images by taking photos of the pantograph slide oblique side.Considering that the measuring plane is not parallel to the imaging plane,and in measuring plane,the two sliding are different.Through the method of perspective correction,the image of the pantograph slide was corrected.According to the corrected pantograph slide picture,using MATLAB application software to preprocess pantograph slide image,including pantograph slide image gray processing,pantograph slide image filtering,pantograph slide image enhancement processing.After image preprocessing,the edge image of the pantograph slide is clearer,and increasing the contrast between the target with the background.(2)Comparing different edge detection algorithms by comparison the detection effects.Firstly,using canny edges detection to detect the edges of pantograph slide.Realize the detection of the wear edge and the bottom edge of the pantograph slide;Secondly,by morphological open and close algorithms process,remove the false edge after edge detection,close the breakpoint;Finally,by refinement obtained continuous and accurate pantograph slide edge.Judging the two edges starting point of the pantograph slide,through image edge tracking locate the edge of pantograph slide,extracting the pixel coordinates of the pantograph slide edge.(3)Since the two ends of the pantograph slide do not contact with the catenary net during work,It can be used as a reference position for imaging judgment.Firstly,selecting the two ends of the pantograph slide to calibrate the pixel equivalent.And calibrating the pixel equivalent of the two pantograph slide respectively in different measurement planes.Secondly,calculating the thickness of the pantograph slide through the ordinate of the pantograph slide edge and the calibration parameters,make the wear curve of the pantograph slide and calculate the lowest point of the pantograph slide wear.Finally,establishing MATLAB GUI simulation system,through the system doing the Measuring test to the pantograph slide thickness.In this thesis,by using MATLAB software to process pantograph slide image data,through the use of perspective correction method to reduce the impact of perspective distortion;In view of the non-linear characteristics of pantograph slide bottom edge,by using the edge tracking method to track and locate the bottom edge of pantograph slide.The algorithm implemented small edge positioning error range in the thickness measurement of the pantograph slide.It is found that the algorithm has a high accuracy and has certain value to the practical application in the field.