NHA12D: A New Pavement Crack Dataset and A Comparison Study of Crack Detection Algorithms
View / Open Files
Conference Name
2022 European Conference on Computing in Construction
Type
Conference Object
This Version
AM
Metadata
Show full item recordCitation
Huang, Z., Chen, w., Al-Tabbaa, A., & Brilakis, I. NHA12D: A New Pavement Crack Dataset and A Comparison Study of Crack Detection Algorithms. 2022 European Conference on Computing in Construction. https://doi.org/10.17863/CAM.83123
Abstract
Crack detection plays a key role in automated pavement
inspection. Although a large number of algorithms have
been developed in recent years to further boost perfor-
mance, there are still remaining challenges in practice, due
to the complexity of pavement images. To further acceler-
ate the development and identify the remaining challenges,
this paper conducts a comparison study to evaluate the per-
formance of the state of the art crack detection algorithms
quantitatively and objectively. A more comprehensive an-
notated pavement crack dataset (NHA12D) that contains
images with different viewpoints and pavements types is
proposed. In the comparison study, crack detection al-
gorithms were trained equally on the largest public crack
dataset collected and evaluated on the proposed dataset
(NHA12D). Overall, the U-Net model with VGG-16 as
backbone has the best all-around performance, but models
generally fail to distinguish cracks from concrete joints,
leading to a high false-positive rate. It also found that
detecting cracks from concrete pavement images still has
huge room for improvement. Dataset for concrete pave-
ment images is also missing in the literature. Future di-
rections in this area include filling the gap for concrete
pavement images and using domain adaptation techniques
to enhance the detection results on unseen datasets.
Sponsorship
National Highways PhD studentship
Embargo Lift Date
2023-04-01
Identifiers
External DOI: https://doi.org/10.17863/CAM.83123
This record's URL: https://www.repository.cam.ac.uk/handle/1810/335687
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.