Proceedings of the 2022 European Conference on Computing in Construction
 |
NHA12D: A new pavement crack dataset and a comparison study of crack detection algorithms
Zhening Huang, Weiwei Chen, Abir Al-Tabbaa, Ioannis BrilakisUniversity of Cambridge, United KingdomDOI: 10.35490/EC3.2022.160Abstract: This paper conducts a comparison study to evaluate the performance of the state of the art crack detection algorithms quantitatively and objectively. A more comprehensive annotated pavement crack dataset (NHA12D) that contains images with different viewpoints and pavements types is proposed. In the comparison study, crack detection algorithms were trained equally on the largest public crack dataset collected and evaluated on the proposed dataset (NHA12D). Results: models generally fail to distinguish cracks from concrete joints. Detecting cracks from concrete pavement images still has huge room for improvement. Domain adaptation algorithms can be used to boost performance on unseen images. Keywords: crack detection, crack dataset, benchmark study, computer visionPaper:EC32022_160