Feature Extraction Using Digital Image Correlation for Pulsed Eddy Current Thermographic ImagesSummaryPulsed Eddy Current Thermography Pulsed Eddy (PEC) is a new method in the field of non-destructive testing and evaluation (NDT&E). Heat diffusion in a sample can be characterized by analyzing a sequence of PEC thermography images, which correspond to defects. This study leverages the capabilities of PEC thermography to obtain quantitative information for angular defect characterization through surface thermal distribution analysis. To perform this analysis, a new technique using digital image correlation (DIC) was proposed to track heat diffusion through sequential PEC thermographic images in a metal sample. The analysis results were used to extract the defect characteristics of the tested sample. These results showed the effectiveness of the proposed technique in providing features that are in good agreement with defect detection and evaluation. Keywords: pulsed eddy current thermography, digital image correlation, feature extraction, quantitative evaluation.1. IntroductionThere is a growing need for efficient, rapid and reliable non-destructive evaluation (NDE) methods and techniques to inspect complex engineering components and structures to recognize possible failure sites. Early detection and quantitative information of a defect prove to be an elementary factor in predicting the lifespan of a component [1]. The pulsed eddy current (PEC) thermography technique has been used for the detection and characterization of defects in conductive materials over a relatively wide area [2]. The technique uses induced eddy currents to heat the material under test...... middle of article ......a conference on NDT, Berlin, Germany (2006).[17] LCS Nunes, DA Castello, CF Matt, PAM dos Santos, “Parameter estimation using digital image correlation and inverse problems”, Marcílio Alves and HS da Costa Mattos, Brazilian Society of Science and Technology mechanical engineering, pp. 432-443, 2007. [18] AJH Hii, CE Hann, JG Chase, EEW Van Houten, "Fast normalized cross-correlation for motion tracking using basis functions", Methods and Computer Programs in Biomedicine, Flight. 82, pp. 144-156, 2006.[19] JC Russ, Model matching and correlation, in: “The Image Processing Handbook”, 2nd ed., CRC Press, Raleigh, pp. 341-346, 1994.[20] C. Eberl, R. Thompson, D. Gianola, “Digital Image Correlation and Tracking in Matlab,” Matlab Central, 2006. http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=12413&objectType =file///).