Abstract
Surface cracks in concrete structures are critical indicators of structural damage and durability. The vision-based methods can automatically extract crack information from images. Standardizing crack identification using image binarization and region classification, is challenging because of the parameters dependence and high time consumption. This paper presents a fast adaptive crack detection algorithm that has an adaptive binarization procedure without any specific parameter and a machine learning-based classification procedure. Firstly, according to the double edge characteristics of cracks, a finite state machine (FSM) operator is designed. The operator searches valleys and hillsides on the grayscale curve, which are the location of candidate cracks. While the image is processed by the operator, the features of crack regions can be computed directly, which composes the crack samples in the manual marked images. Secondly, a random forest classifier is trained and tested by the samples. Crack detection experiments on concrete components prove that the average detection sensitivity is over 93%, and the time complexity is extremely low that the average processing time of megapixel images is 95 ms.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.