Research On Rail Crack Detection Based On Infrared Image Processing

As a key component of railway operation,rail is prone to damage due to the dual effects of the external pressure and natural factors.It is the precondition to ensure the stable operation of the train in time to find the damaged rail and replace it,which is very important to ensure the safety of people’s lives and reduce economic losses.In recent years,there has been a lot of scientific research on the crack detection technology of rail at home and abroad.Ultrasonic testing is currently the most widely used methods to detect cracks in railway,but the traditional ultrasonic testing method has a blind spot in the rail crack detection,it can’t detect the cracks on both sides of the rail foot.Aiming at the deficiency of the ultrasonic testing method,the infrared thermal wave nondestructive testing technology is used to detect the cracks on both sides of the rail foot.Based on infrared thermal wave nondestructive testing technology,the rail bottom crack detection platform is built.Due to the low precision and low contrast of the infrared image of the rail bottom crack,a new infrared image enhancement algorithm based on the combination of high pass filtering and histogram equalization is proposed to improve the precision of the rail bottom crack geometry.The main contents are below.First of all,the numerical simulation of infrared thermal wave nondestructive testing of rail bottom crack is carried out in the ANSYS finite element analysis software,the simulation process mainly includes the following three parts.A three-dimensional solid model with cracks is established in the pre-processing module of ANSYS,choosing the appropriate meshing method for rail bottom crack model.The heat flux is simulated on the surface of the rail foot,and the results are analyzed and solved by ANSYS solver.In the post processing module of ANSYS,the temperature distribution of the rail foot surface is generated by the ANSYS universal postprocessor,the temperature versus time curve of the key point is generated by the ANSYS time history processor.Secondly,a rail bottom crack detection platform based on infrared thermal wave nondestructive testing technology was set up,the construction of the experimental platform mainly includes the following three parts.According to the requirements of the infrared thermal wave nondestructive test of rail bottom crack experiment,a suitable thermal excitation source is selected.Active heating on the rail surface of rail foot with the selected heat source,the distribution of temperature field on the surface of the rail foot caused by the thermal wave propagation in the rail is recorded by the thermal imager,and then the infrared thermal image of the rail bottom crack is collected.The infrared image of the rail bottom crack is transmitted to the computer,and the temperature information contained in the image is analyzed by the online analysis software.Finally,the infrared image of the rail bottom crack was processed.Aiming at the deficiency of traditional image processing methods for the infrared image processing of rail bottom cracks,a new algorithm of infrared image enhancement based on high pass filtering and histogram equalization is proposed.Compared with other algorithms,the algorithm can meet the demand of rail bottom crack detection,and improve the accuracy of the calculation of rail bottom crack geometry.