Quantitative Delamination Imaging in CFRP Composites Using Lamb Waves: Accounting for Material Uncertainty via the FBP Method

Material property variability in carbon fiber-reinforced polymer composites is a major source of uncertainty in quantitative Lamb wave-based delamination imaging. Even minor deviations in elastic properties can alter dispersion characteristics and wave propagation behavior, thereby reducing the reliability of imaging-based assessments. To systematically investigate this effect, the present study examines the influence of subtle material variations on Lamb wave responses through combined numerical modeling and finite element simulations. Time-of-flight features at the excitation frequency are extracted using a continuous wavelet transform with Morlet wavelets, enabling robust identification of mode-dependent arrival information. Within a finite element framework, A0 mode propagation in the presence of delamination is simulated, and damage imaging is subsequently carried out using a filtered back-projection algorithm. To improve boundary definition and support quantitative characterization, morphological filtering is combined with Canny edge detection to refine the reconstructed damage contours. The results show that the proposed filtered back-projection-based imaging framework provides clear and accurate visualization of delamination while exhibiting reduced sensitivity to material property uncertainty. Delaminations with different geometric characteristics are consistently identified, demonstrating enhanced robustness against material variability. These findings confirm the effectiveness of the proposed Lamb wave-based approach for reliable, quantitative delamination imaging in composite structures.