Detection and Measurements of Cracks in Axially Loaded Tension RC Members by Image Processing Technique

Concrete cracking is very common and appeared in all types of concrete structures. Detection of concrete crack initiation and measurement of their characteristics has been a part of structural monitoring and maintenance program. Though manual inspection of cracks in the concrete structure is the oldest and still an accepted method of crack measurements, a more accurate technique for crack detection and measurement, applicable for all kind of concrete structures, is required. Automatic crack identification and crack width measurement by the image analysis method has been gaining popularity due to its versatile applicability, more accuracy and safety standard in difficult situations. In this context, the image analysis method is adopted in laboratory testing of the concrete prism with reinforcement under axial tension to identify and measure cracks in concrete prism samples. In this research work, a crack quantification method based on 2D image analysis is utilized. Raw colour images (RGB images) were obtained by means of a camera for cracks formed due to axial tension in reinforcement. Processing of RGB images with sufficient information will provide the dimension of crack. For all the image post-processing, an open source ImageJ software was utilized and characteristics of cracks were automatically determined by Ridge Detection Plugin. The results of image processing were also verified with the results from microscopic measurement of cracks in reinforced concrete prism samples