Quantitative analysis of thermographic data through different processing techniques

Tamborrino, Rosanna; Politecnico di Bari; Italy

Cavallo, P.; Diagnostic Engineering Solutions srl; Italy
D`Accardi, E.; DMMM - Politecnico di Bari- Viale Japigia; Italy
Galietti, U.; DMMM - Politecnico di Bari- Viale Japigia; Italy
Palumbo, D.; DMMM - Politecnico di Bari- Viale Japigia; Italy
Tamborrino, R.; DMMM - Politecnico di Bari- Viale Japigia; Italy

ID: ECNDT-0564-2018
Session: Thermography and Thermosonics Techniques 1
Room: H1
Date: 2018-06-15
Time: 09:40 - 10:00

Thermography is commonly used as non-destructive technique for evaluating defects within materials and components. However, raw thermal imaging data are usually not suitable for quantitative evaluation of defects and often the choice of a particular processing technique is too limited and/or inappropriate to an accurate detection of defects. It was necessary to correctly process the raw thermal data acquired to obtain a series of satisfactory results for a quantitative evaluation of sizes and depths of the discontinuities.
In this work, various algorithms have been implemented to elaborate raw thermal data and to characterize the defects.
Starting from the same thermographic tests carried out on an aluminium specimen and on a composite specimen, the algorithms, focused on the evaluation of several parameters, have been applied on the same thermographic sequence and the results have been compared. The algorithms used have been: Pulsed Phase Thermography (PPT), Slope, Correlation Coefficient (R2), Thermal Signal Reconstruction (TSR), Principal Component Analysis (PCT), Amplitude (A) and Phase (P). By analysing the thermographic images obtained using the different approaches, it was possible to focus on the advantages, disadvantages and sensitivity of each thermographic algorithm implemented.