In this study, three different CFRP specimens with internal artificial delaminations of various sizes and located at different depths were investigated by means of Pulsed Thermography (PT) under laboratory conditions. The three CFRP panels, having the same thickness and defects characteristics but with a different shape (planar, trapezoid and curved), were assessed after applying various signal processing tools on the acquired thermal data (i.e. Thermographic Signal Reconstruction, Pulsed Phase Thermography and Principal Component Thermography). The effectiveness of the above processing tools was initially evaluated in a qualitative manner, comparing the imaging outputs and the information retrieval in terms of defect detectability enhancement and noise reduction. Simultaneously, the produced defect detectability was evaluated through Signal-to-Noise Ratio (SNR) computations, quantifying the image quality and the intensity contrast produced between the defected area and the adjacent background area of the test panel. From the results of this study, it can be concluded that the implementation of PT along with the application of advanced signal processing algorithms can be a useful technique for NDT assessment, providing enhanced qualitative information. Nevertheless, SNR analysis showed that despite the enhanced visibility resulting from these algorithms, these can be properly applied in order to retrieve the best possible information according to the user’s demands.