1/1
2 files

SSD and YOLOv4 Model for Structural Detail Detection

software
posted on 07.10.2021, 15:41 authored by Eric BianchiEric Bianchi, Matthew Hebdon

This file includes two models trained on COCO-Bridge 2021+. The first model is SSD. This model expects images to be 300x300. The second model is YOLOv4. YOLOv4 expects images to be 320x320. These object detection models find four structural bridge details: [bearings, cover plate terminations, gusset plate connections, and out of plane stiffeners]. This model was trained using the COCO-Bridge 2021+ dataset (DOI: 10.7294/16624495). For a confidence threshold of 25% and an IoU threshold of 50%, the SSD (with augmentation) and YOLOv4 model achieved mean average precision (mAP) scores of 60% and 88%, respectfully. The training and testing code for YOLOv4 and SSD can be found by visiting the GitHub repository pages referenced in the journal article.


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

Funding

National Science Foundation Grant No. IIS-1840044

History

Publisher

University Libraries, Virginia Tech

Language

English (US)

Location

Virginia Tech