Defect detection and evaluation of ultrasonic infrared thermography for aerospace CFRP composites

The ultrasonic infrared thermography Non-destructive Testing is introduced for detecting the impact damage of a CFRP specimen for Unmanned Aerial Vehicles. The characteristics of thermal images with damage are particularly analyzed. A Local Binary Fitting (LBF) model based on a non-Gaussian kernel function is used to segment the defect edge. In view of the discontinuity of defect in thermal images due to multilayered structure of composite materials, defect merging algorithms are proposed including time domain and space domain methods by using a few thermal images, and the defect geometric distortion during camera imaging is also compensated. The defect in the composite material can be quantitatively analyzed after the defect reconstruction. The experimental result has shown that the proposed algorithm can effectively detect and evaluate the impact damage of thermal images and the accuracy of quantitative assessment is correspondingly increased.