Image fusion of ultrasonic and thermography for delamination detection of CFRP based on data post-processing methods

Delamination is crucial damage in carbon fibre reinforced polymer (CFRP) composites. Various non-destructive testing methods could detect the delamination with different merits. Ultrasonic can penetrate CFRP but has the dead zone effect, whereas thermography can reveal shallow defect but is insensitive to deep delamination. Therefore, in this work, the results of ultrasonic C-scan and locked-in thermography were fused to inspect the delamination. The ultrasonic C-scan images were processed by 6 dB drop criteria to quantify defect area. Principal component thermography, thermographic signal reconstruction and phase extraction were implemented on the thermographic images. C-scan images and thermographic images were fused using the proposed fusion algorithm that combines the evidence theory and minimum values. The performances of fused images were evaluated by the index of signal-to-noise ratio. The results show that the C-scan image fused with the thermographic phase image can significantly improve the signal-to-noise ratio of shallow defect detection and reduce the error with the actual defect area.