Defect extraction and quantitative analysis of composite materials based on infrared detection

Carbon fiber reinforced polymer (CFRP) composite materials have been widely used in the aircraft industry due to their excellent properties. Defect damage in the composite materials is unavoidable produced during the manufacturing process and the in-service procedure. It is of great significance to use accurate and efficient non-destructive testing(NDT) technology to detect the defect damage of composite materials for ensuring the safe operation of aircraft in service. In this paper, the low-speed impact damage defects in civil aircraft composites were detected by pulsed thermography. Aiming at the shortcomings of the original thermal wave image, such as blurred defect details and edges, and low contrast, Otsu double-threshold method is used to segment and extract defects. By adding image enhancement based on histogram equalization algorithm and background suppression based on Wiener filtering, the phenomenon of false segmentation and noise residual is avoided. In addition, quantum genetic algorithm is introduced to improve the efficiency of the algorithm. The results show that the image processing method designed in this paper can better enhance the display of defects in infrared image and suppress background interference and noise, making the defect extraction more accurate and fast. Compared with the results of ultrasonic C-scan, the relative identification error of defect extraction results is less than 5%, which is satisfied with the damage detection requirements of carbon fiber reinforced composite in in-service aircraft.