Virginia Tech
3 files

Bearing Condition State Classification Dataset

posted on 2021-10-05, 17:22 authored by Eric Bianchi, Matthew Hebdon

This is a dataset of structural bridge bearings. The bearings have been annotated using the American Association of State Highway and Transportation Officials (AASHTO) bridge inspection condition state guidelines and Bridge Inspector's Reference Manual (BIRM). The authors have included annotation guidelines and provided examples and explanation for bearings and their respective condition state assessment. There are a total of 947 images of bearings included in the dataset. The image size is 300x300. The bearing images were obtained from the COCO-Bridge-2021+ (Bianchi) dataset for structural detail detection. The data was split 10% testing, 90% training. After training with the EfficientNet B3 model (DOI: 10.7294/16628698), we were able to obtain an F1 score of 86.4%. More details of the training, the results, the dataset, and the code may be referenced in the journal article. The GitHub repository information may be found in the journal article.

If you are using the dataset in your work, please include both the journal article and the dataset citation.


National Science Foundation Grant No. IIS-1840044



University Libraries, Virginia Tech


  • English (US)


Virginia Tech