Abstract
This paper shows how 2D digital image correlation (2D DIC) and region-based convolutional neural network (R-CNN) can be combined for image-based automated monitoring and assessment of surface crack development of concrete structural elements during laboratory quasi-static tests. In the presented approach, the 2D DIC-based monitoring enables estimation of deformation fields on the surface of the concrete element and measurements of crack width. Moreover, the R-CNN model provides unmanned simultaneous detection and localization of multiple cracks in the images. The results show that the automatic monitoring and evaluation of crack development in concrete structural elements is possible with high accuracy and reliability.
Highlights
Much of the important concrete structures that are in use today were erected several years ago and they are close to their design life [1]
We present the results of application of digital image correlation and region-based convolutional neural networks in evaluation of surface cracks in the crane runway beam during experiments described in the previous section
We have developed a computer vision system for automatic assessment of crack development visible on the surface of a concrete structural element during laboratory quasi-static tests
Summary
Much of the important concrete structures that are in use today were erected several years ago and they are close to their design life [1]. Condition assessment of a concrete structural element involves monitoring of displacements and evaluation of crack development visible on the surface of the element during laboratory mechanical tests. Monitoring is often done using wired contact sensors such as linear-variable-differential transformers (LVDTs). These sensors are often difficult to install and maintain. Crack evaluation typically involves visual inspection by trained staff and simple measuring tools such as a Brinell magnifier. Such inspection methods can be expensive, dangerous and time-consuming. Computer vision methods allow for fully automated extraction of important information from digital images [2,3]. Various methods leveraging image-processing techniques and computer vision methods such as digital image correlation and convolutional neural networks have been developed and applied in past decades [4,5,6,7,8,9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
