Defect Detection Of Material Surface Based On Reflective Laser Thermal Imaging

Non-destructive testing is widely used in industries such as industrial production,transportation,energy and infrastructure.When a defect occurs in the material test piece,it is easy to cause stress concentration in the cross section,resulting in deterioration of serviceability and safety problems.Therefore,regular non-destructive testing is of great significance.At present,there are a variety of non-destructive testing methods.Among them,infrared thermal imaging is a high-speed,non-contact testing technology,and it can realize the visualization of defects at the same time.In this paper,for the problem of infrared thermal imaging nondestructive testing,a detection system based on laser excitation is constructed and experiments are carried out.The main work is as follows:(1)The research status of infrared thermal imaging nondestructive testing is reviewed,and the theoretical basis of pulsed laser thermal imaging nondestructive testing technology is analyzed;The effects of material surface defects on the heat transfer process under laser excitation were studied,and the simulation analysis was performed using the finite element analysis software ANSYS;the composition of the laser-excited infrared thermal imaging nondestructive detection system was discussed,the various modules of the system were introduced,and the experiment was built The system was verified by experiments.(2)In order to remove the noise of the infrared image,the type of noise in the image is analyzed first,and the evaluation method for the quality of the infrared image is introduced;the image restoration method is used to remove noise,improve the clarity,and complete the restoration based on the edge method Calculation of parameters;Wiener filtering and blind convolution are used to complete the restoration process,and the advantages and disadvantages of the two algorithms are compared through experiments.(3)In the experiment of laser spot thermal imaging defect detection,a single feature extraction method is proposed for the defects detected in a single time;the information of different boundaries of the defect is extracted through multiple detection;the image is determined by the maximum between-class variance method Segmentation threshold and semithreshold segmentation;based on image fusion and boundary sharpening method to achieve defect feature extraction.The research in this paper shows that: defects will hinder the propagation of heat flow inside the material,and verification will be completed through the combination of simulation and experiment;the image restoration algorithm can effectively remove image noise and improve image clarity,and Wiener filtering is better The restoration effect of the image is better;through the feature extraction algorithm,the background is removed,and the integrity of the defect feature is retained,and the quality is improved.