PROJECT TITLE :

Automated Crack Detection on Concrete Bridges

ABSTRACT:

Detection of cracks on bridge decks may be a vital task for maintaining the structural health and reliability of concrete bridges. Robotic imaging can be used to get bridge surface image sets for automated on-website analysis. We have a tendency to gift a completely unique automated crack detection algorithm, the STRUM (spatially tuned sturdy multifeature) classifier, and demonstrate results on real bridge data using a state-of-the-art robotic bridge scanning system. By using Machine Learning classification, we tend to eliminate the need for manually tuning threshold parameters. The algorithm uses sturdy curve fitting to spatially localize potential crack regions even in the presence of noise. Multiple visual options that are spatially tuned to those regions are computed. Feature computation includes examining the size-house of the local feature so as to represent the information and also the unknown salient scale of the crack. The classification results are obtained with real bridge data from tons of crack regions over two bridges. This comprehensive analysis shows a peak STRUM classifier performance of 95p.c compared with sixty ninepercent accuracy from a additional typical image-primarily based approach. In order to create a composite world view of a giant bridge span, a picture sequence from the robot is aligned computationally to form never-ending mosaic. A crack density map for the bridge mosaic provides a computational description along with a global view of the spatial patterns of bridge deck cracking. The bridges surveyed for data collection and testing include Long-Term Bridge Performance program's (LTBP) pilot project bridges at Haymarket, VA, USA, and Sacramento, CA, USA.