Objective Infrared lock-in thermography is an effective non-contact method for detecting internal delamination defects in carbon fiber reinforced polymer (CFRP) composites, while its detection capability for deep or small-scale delamination is constrained by thermal diffusion length and blind-frequency phenomena. In this paper, a non-contact nondestructive testing and evaluation approach that integrates infrared lock-in thermography with air-coupled ultrasound is proposed. In this method, infrared lock-in thermography is employed to identify near-surface delamination defects in CFRP composites, while air-coupled ultrasonic testing is utilized to detect deeper interfacial delamination defects. More specifically, a standardized preprocessing workflow is first established for both thermal image sequences and air-coupled ultrasonic data, where adaptive background suppression and sub-pixel geometric registration are incorporated to effectively mitigate inherent spatial and intensity-scale mismatches. Second, after extracting efficiently the thermal and ultrasonic features of CFRP delamination defects, a multidimensional quantitative evaluation framework is constructed, encompassing the metrics for signal quality, geometric accuracy, and spatial–topological consistency, moreover, the error analysis of the individual inspection modalities is also conducted. Subsequently, the obtained thermal and ultrasonic features are effectively fused using a Laplacian pyramid fusion algorithm, and the impact of feature selection, feature weight, and fusion algorithms are also analyzed in-depth. Finally, a CFRP specimens contains 16 flat-bottomed holes at varying diameters and depths are designed, and the corresponding infrared lock-in thermography and air-coupled ultrasonic testing experiments are also performed to validate the proposed method. Results reveal that, compared to either infrared lock-in thermography or air-coupled ultrasonic testing alone, the proposed method can significantly improve the detection effect of small-size delamination defects in CFRP composites, and increases the signal-to-noise ratio by up to 15.53 dB and the absolute contrast-noise ratio by up to 3.45.

Methods This paper takes advantage of the sensitivity of infrared lock-in thermography to near-surface interface delamination defects and the sensitivity of air-coupled ultrasonic testing to internal deep interface delamination defects, and proposes a non-contact non-destructive detection and evaluation method for internal interface delamination defects in CFRP laminates by integrating infrared lock-in thermography and air-coupled ultrasonic testing. To verify the effectiveness of the proposed method, a CFRP specimen contains 16 flat-bottom holes at varying diameters and depths was designed (Fig.1), and corresponding infrared lock-in thermography and air-coupled ultrasonic testing were carried out (Figs.2 and 3). Finally, the effects of feature selection, feature weights, and data fusion algorithms were compared and analyzed (Fig.7).

Results and Discussions Through a series of experiments on infrared lock-in thermography and air-coupled ultrasonic testing, the superior performance of the proposed method in the high-precision NDT and evaluation of CFRP delamination defects is verified. To highlight small-size delamination defects in CFRP laminates, the thermal phase feature at the excitation frequency of 0.01 Hz and the ultrasonic feature image at the inspection frequency of 400 kHz are selected. The effects of different weight combination and data fusion algorithms on the detection effect are investigated in depth. The optimal feature weight combination and data fusion algorithms are given in this study, i.e., the optimal weight combination for thermal wave phase features and ultrasonic feature images are respectively set to 0.3 and 0.7, and the proposed detection method based on blending Alpha fusion algorithm can significantly improve the detection effect of small-size delamination defects in CFRP laminates.

Conclusions This paper first proposes a non-contact nondestructive testing and evaluation method for CFRP delamination defects by infrared lock-in thermography fusing air-coupled ultrasonic testing. A systematic study on feature extraction, feature registration, feature selection, data fusion, and quantitative evaluation is conducted. Subsequently, to validate the effectiveness of the proposed method, CFRP specimens contains artificially delamination defects at varying diameters and depths were fabricated, and the corresponding infrared lock-in thermography and air-coupled ultrasonic testing were also carried out. Finally, a comparative study of different feature weight combinations and data fusion algorithms was performed, and the fusion results were also quantitatively evaluated by using these metrics including signal-to-noise ratio, size deviation, and hit rate. The results indicate that, compared to single detection modality using either infrared lock-in thermography or air-coupled ultrasonic testing, the proposed detection method based on Laplacian pyramid fusion algorithm significantly improves the detection effect of CFRP delamination defects. Specifically, the proposed method can significantly improve the detection effect of small-size delamination defects in CFRP composites, and increases the signal-to-noise ratio by up to 15.53 dB and the absolute contrast-noise ratio by up to 3.45. Furthermore, the weights assigned to thermal wave phase features or ultrasonic detection features have a significant impact on the detection results for CFRP delamination defects. When the weights for thermal wave phase features and ultrasonic feature images are respectively set to 0.3 and 0.7, —giving dominance to air-coupled ultrasonic features— the CFRP delamination defects can be clearly observed, but its texture details are lacking. Conversely, when the weights are respectively set to 0.7 and 0.3— prioritizing thermal wave phase features— the fused detection results not only reveal the interfacial delamination defects but also retain rich texture details.