The Image Processing Technique of Defect Detection in Metal Materials Using Active Infrared Thermography

Abstract

Technology on active infrared thermography has been widely used in the field of non-destructive testing. The defects produce an uneven heat dissipation, and an infrared camera captures this thermal reaction. The phase image provides an outstanding invisible flaws description on sequences of images captured. During the data acquisition phase, the result might be affected by image noise resulting from the emissivity on the specimen surface and also by the non-uniform heating process. For this study, an infrared camera of medium wave infrared range was analysed. The heating process is done by heating the aluminium plate in an oven for 5–15 min. These specimens contain twelve numbers of flat-bottomed circular holes, from sample upper surface with different artificial depth. All of the artificial defect samples were manufactured by the milling machining process. The defect detection and characterisation used several data and different analysis methods were used in the pre-processing for quantitative analysis. The infrared test method on non-destructive thermography capability was increased since the final result appears clearer and can be used for future studies. The temperature data used the image filtering and image segmentation process for better visualisation of defects. The correlation image and the correlation image phase show promising results. The experimental setup and comparative results were analysed in detail in this document. As a conclusion, the image enhancement method has a significant influence on the defect detection result.