Regularization techniques for thermographic image reconstruction
Sprache des Vortragstitels:
Englisch
Original Tagungtitel:
5th NDTonAIR Training Event & Thermography Workshop
Sprache des Tagungstitel:
Englisch
Original Kurzfassung:
Thermography is a non-destructive testing technique where the specimen is heated and the corresponding temperature evolution is measured on the sample surface. This is used to reconstruct the heat distribution inside the material. We focus on thermographic imaging, in which multiple 1D surface measurements are combined to reconstruct the original heat distribution in the cross section of the specimen. In this talk, we apply different regularization techniques to solve the image reconstruction problem in question. We compare the performance of these methods in terms of estimation accuracy and computational complexity. As a case study, we consider the 2D thermographic image reconstruction problem by utilizing the so-called virtual wave concept, and demonstrate the results in MatLab.
Sprache der Kurzfassung:
Englisch
Englische Kurzfassung:
Thermography is a non-destructive testing technique where the specimen is heated and the corresponding temperature evolution is measured on the sample surface. This is used to reconstruct the heat distribution inside the material. We focus on thermographic imaging, in which multiple 1D surface measurements are combined to reconstruct the original heat distribution in the cross section of the specimen. In this talk, we apply different regularization techniques to solve the image reconstruction problem in question. We compare the performance of these methods in terms of estimation accuracy and computational complexity. As a case study, we consider the 2D thermographic image reconstruction problem by utilizing the so-called virtual wave concept, and demonstrate the results in MatLab.
Vortragstyp:
Hauptvortrag / Eingeladener Vortrag auf einer Tagung