Research On The Surface Defect Detection System Of Bearing Cylindrical Roller Based On Machine Vision

Bearings are the core key components in railway train transportation,and cylindrical rollers,as an important part of roller bearings,also play an indispensable role.In the process of railway transportation,under the condition of high speed and heavy load,cylindrical rollers often have various defects on their surface,leading to potential safety hazards in train driving,so periodic inspection and repair of bearings are required.In recent years,the application of machine vision in the defect inspection of bearing rollers has become more and more extensive,and it has become the mainstream of defect inspection.Based on the existing roller surface inspection system,this thesis combines theoretical analysis,simulation verification and physical experiments to design a set of machine vision bearing cylindrical roller surface defect inspection system.The main research contents are as follows:Categorize and analyze the surface defects of cylindrical rollers.Determine the conveying and unfolding methods of cylindrical rollers,and select hardware models for cylindrical roller surface defect inspection systems.Establish cylindrical rollers dynamic model,analyze the vibration response of the cylindrical roller conveying process.Complete the overall scheme design of the cylindrical roller surface defect inspection system.Arrange the light source reasonably.Complete the formulation of the cylindrical roller image acquisition plan.Determine the influencing factors that affect the image defect of the cylindrical roller acquisition,determine the evaluation index for evaluating the image defect of the cylindrical roller.Build an experimental platform to influence the image acquisition defect of the cylindrical roller Factors are analyzed,the best conditions for image acquisition of cylindrical rollers are obtained through experiments,and the design of the cylindrical roller image acquisition system is completed.The acquired roller images are preprocessed to facilitate subsequent defect identification.Extract the area of interest from the original image of the cylindrical roller.Obtain the images of the end face and cylinder of the roller,and reasonably design the template matching algorithm to speed up the algorithm running,Process the end face and cylinder images of the cylindrical roller to obtain the surface defects of the cylindrical roller The visual characteristics of the roller are established,the defect identification process is established.Design testing system software interface and the surface defect detection of the roller is completed.This thesis designs the surface defect inspection system of cylindrical roller bearings based on machine vision,realizes the automatic inspection of cylindrical roller surface defect,solves the problem that cylindrical rollers are not easy to expand while traveling,and provides efficient and feasible defect detection algorithms,which can be used for cylindrical rollers.The follow-up study of the surface defect inspection system provides a theoretical basis.