Abstract
Infrared thermographic nondestructive testing is a proven technology for material inspection and characterization that is based on the analysis of the dynamic thermal behavior of specimens induced by an intentional and controlled external stimulation. The analysis and review of the image sequences resulting from the inspections is time consuming, which cancels out the advantage provided by the quick inspection process. The RGB projection method has been recently proposed to overcome this limitation. This method synthesizes long sequences into only one true color image, selecting the three most representative images from the original sequence. This study advances the development of this methodology by automating the selection of the channels to be included in the RGB image, making the process objective. Seven different criteria based on different metrics are proposed to select the channels. They are analyzed and evaluated by theoretical and real thermographic tests obtaining RGB images that contain up to 99.6 % of the defects detected in the original thermographic sequences. Visualization improvement is also provided to the RGB image by image processing algorithms.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call