The Canny shape-reconstruction algorithm is combined with the thermographic signal reconstruction (TSR) technique for the characterization of delamination defects in planar multilayer structures. The adopted approach consists in fitting the raw thermographic images to a low-degree polynomial in the log-log representation of the time axis, according to TSR technique, and then apply the Canny algorithm to the reconstruction of the original signals. The proposed juxtaposition of the two treatments benefits from the inherent signal de-noising and compression of the TSR algorithm, thus enhancing the performance of the overall reconstruction process. The proposed approach has been applied using synthetic signals obtained via a 3D semi-analytical simulation code and the obtained reconstructed images have been compared with the (known) delamination geometry demonstrating satisfactory results.