GUI development for automatic defect visualization and size estimation using image analysis processing

In this paper, the authors propose an algorithm and GUI for automatic defect detection and defect size estimation using scanned images from non-destructive testing systems. The test specimen was fabricated by laser-engraving a triangle and square shape on non-transparent acrylic plate. Non-destructive testing was performed using a terahertz imaging system. The algorithm for the defect detection and the size estimation was considered a binary image transformation algorithm to enable data from various inspection systems that are in non-standardized data format. The region of interest (ROI) function was applied to perform the algorithm within an assigned mask area on the image. As a result, the accuracy of the proposed algorithm was improved. In addition, the developed GUI is configured to enable multi-ROIs to analyze the multi-defects simultaneously.