Research On Imaging System Of Track Defect Detection Based On Machine Vision

Track detection can provide defect data for railway maintenance,which is becoming one of the most necessary means to eliminate safety hazards of railways.Because of its fast speed,high precision and high reliability,the detection of track defects based on machine vision is becoming a hot spot in the research of track detection.Because the stability and quality of the original image acquisition in the machine vision system will directly affect the results of the image processing,an appropriate orbital defect imaging system can reduce the influence of the external environment on the imaging system,improve the speed of image processing,and reduce the missed rate or the rate of failure,the study of the orbital defect imaging system is very important.To improve the imaging quality of track detection system based on machine vision,this paper regards the purpose to improve the image quality as the starting point.In this paper,the effects of imaging methods on surface defects,the characteristics of track surface defects and the machine vision imaging model are studied,and the conclusion of the model is verified by experiments.The main research and contributions of this thesis are as follows:In order to achieve the goal of high quality image acquisition on the rail surface,an optical imaging model of surface defect detection is established.By analyzing the influence of system vibration on image quality,the model is extended to the mathematical model of machine vision system based on linear CCD camera to analyze the relationship between the depth of the image and the image gray scale,the imaging mode,and the relationship between the illumination angle and the image gray scale,the contrast.Then the paper concludes that when the angle of illumination is 45 'the quality of the picture is higher than that of other angles,and the correctness of the conclusion in the model is proved by the experiment.Aiming at the problem of concealment of rail surface defects by strong reflection of track surface,the polarized light is introduced into the orbital defect detection,and the optical intensity mathematical model of machine vision detection is established.The gray value of the image is compared with the angle of the polarizer.The relationship between polarized light and strong reflection light is analyzed.Experiments show that by using the polarized light,the strong reflection of light band is effectively suppressed,and that the quality of the track image is improved compared with the original system.In order to analyze the contrast ratio of the images collected by different imaging systems from the objective point of view,and to preserve the characteristics of the human eye to judge the image contrast,introduce the human eye's interest area and the non-interest area weight coefficient,and propose a HVS-free reference image contrast And then the method is used to evaluate the images collected by different imaging systems,and the contrast degree is evaluated objectively,which indicates the advantages and disadvantages of the imaging results.The effect of imaging system optimization is proved by objective quantitative evaluation.