Real Time Detection of Defects in GFRP Bridge Decks Using Infrared Thermography

Abstract

This work is aimed at building a real time system to detect subsurface defects in GFRP bridge decks using infrared thermography. The issues addressed are: (a) development of a real time defect detection system, and (b) image mosaicking to build a composite image map. In the tests conducted, a turn key system was built in Matlab environment using the FLIR SDK to acquire image from the ThermaCAM S60 infrared camera. The images were then analyzed by defect detection algorithms. Efforts were made to minimize the time to detect defects in a captured image. In the second phase, image mosaicking was used to build a “composite image” that combines all the infrared images to form a single image. The location of defects in the “composite image” leads to a system that will be able to point out defects in the bridge as a whole. The study creates a base that can be used for real time defect detection in GFRP bridge decks.