Ageing masonry arch bridges and viaducts form a significant proportion of the transport infrastructure in the UK and Europe. Their assessment and asset management can be highly challenging, often relying on manual inspections that may introduce subjectivity and uncertainty regarding structural condition. While automated approaches based on machine learning, such as damage classifiers, may offer greater objectivity, they are also limited by the quality of training data used in their creation. This has historically been a limiting factor in the development of such methods for masonry structures. This paper presents two datasets of rich image and point cloud data obtained from thirteen masonry bridges and viaducts in the UK, presenting in a range of structural conditions. A subset of the images have been annotated to show mortar joints and common damage types. It is hoped that these datasets, the most comprehensive of their kind currently available, will be beneficial to a wide range of masonry arch bridge research.
Cocking et al. (Tue,) studied this question.