Date of Award
6-2005
Degree Name
Master of Science in Engineering
Department
Electrical and Computer Engineering
First Advisor
Dr. Ikhlas Abdel-Qadar
Second Advisor
Dr. Osama Abudayyeh
Third Advisor
Dr. Sherif Yehia
Fourth Advisor
Dr. Norali Pernalete
Access Setting
Masters Thesis-Open Access
Abstract
Infrared Thermogarphy (IR) is used in this project to detect subsurface conditions in concrete specimen for nondestructive testing and evaluations. This thesis goal is to bridge the gap between Infrared Testing and image analysis by developing an algorithm based on the region growing approach to segment the images and identify the voids without human interference or prior knowledge of the conditions. The segmentation algorithm starts with seed points as the hottest pixels in the image, and then regions are grown based on a neighborhood comparison criterion. The algorithm was tested on images collected from concrete specimen containing various man-made defects and also on defect-free model. The experimental work successfully identified defects from the surrounding bulk material and all the results indicate that Thermography can detect hidden defects up to 3 inches below surface.
Recommended Citation
Yohali, Solange, "Segmentation of Thermal Images for Non-Destructive Applications" (2005). Masters Theses. 4746.
https://scholarworks.wmich.edu/masters_theses/4746