A Research On Defect Detection Of Large-scale Composite Materials Based On Infrared Thermal Wave Technology

In recent years,the proportion of large-scale high-performance composite materials used in the aerospace field has greatly increased due to their excellent performance.Therefore,the research and development of non-contact non-destructive testing technology for defects in large-scale aerospace composite materials is of great significance.This paper combines the detection requirements of in-situ,external field,and overall defects of large-size composite materials,and uses infrared thermal wave imaging technology to collect the temperature field change information of composite materials under thermal excitation to obtain infrared image sequences.We use the temperature in the infrared image sequence The changing characteristics and testing requirements of the composite materials have finally realized the presentation of the complete testing results of large-size composite materials and the quantitative detection of defects.On this basis,the main research contents of this article are as follows:First,for defect recognition in infrared image sequences,a large-scale composite material defect feature reconstruction algorithm based on infrared thermal wave technology is proposed.The automatic segmentation method is applied to realize infrared image sequence block and variable interval search to obtain the corresponding transient thermal response curve(TTR)data set,and design the KG-EM algorithm to classify the TTR data set,obtain the transient thermal response curve with typical temperature change characteristics,and use the curve to reconstruct to obtain Infrared reconstructed image highlighting defect features.This method can extract the main features of the infrared image sequence,and at the same time use the main features to reconstruct the infrared reconstructed image.Visual imaging of the overall defect distribution of large-size inspection specimens.Defect detection and detection of composite material surface defects and sub-surface defects.Then,image stitching is performed on the infrared reconstructed images that characterize the visual imaging results of the surface and internal defects of the composite material.Since the image taken by the thermal imaging camera only covers a small detection area,the overall defect detection result cannot be obtained by a single inspection of a large-size sample.Using the stitching algorithm,the incomplete defect area in a single reconstructed image can be directly generated by stitching complete defect imaging.Therefore,this paper designs a large-scale composite defect reconstruction image mosaic method based on image processing.The SURF-BRISK algorithm is used to quickly extract accurate and effective infrared reconstructed image feature points to construct an image conversion model to achieve visual imaging of the overall defect distribution of large-scale inspection specimens.Finally,a large-scale composite material defect quantitative identification method is designed.This method considers that the object of defect feature extraction is the visual imaging of the overall defect distribution and uses the infrared image sequence of the image source to help obtain the quantitative information of the defect.First,the defect feature extraction algorithm is used for the defects in the infrared reconstructed mosaic image.The registration relationship of the stitched images is used to obtain the transient thermal response curve in the image sequence corresponding to the defect feature.The similarity measurement method of transient thermal response curve is designed to determine the thermal diffusion area of the infrared reconstructed mosaic image defect;finally,the number of pixels in the actual area of the characteristic defect is obtained to realize the quantitative work of the defect detection of large-size composite materials.