Traffic Signs Detection Europe Computer Vision Project
Dataset structure
The dataset contains 55 classes of unique traffic signs. There are 4 main categories:
- forb
- info
- mand
- warn
Each class has one of these prefixes in it's name indicating the category of the traffic sign: forbidden, informational, mandatorry and warning, respectively.
Each image is of size 640x640px. If a traffic sign was recognizable it was labeled but if I could barely tell what the sign was it was not labeled, no matter the distance to the traffic sign. The dataset is not perfectly balanced due to the frequency of traffic signs. Naturally some signs appear more than others. There are some images with no labels, those were collected when models would get false positive detections.
Data collection
Most of the images were collected from Google Maps, Romania, and manually labeled. There's a small part of images that were automatically collected from YouTube videos, by a trained model, and then manually checked (around 150).
Trained Model API
This project has a trained model available that you can try in your browser and use to get predictions via our Hosted Inference API and other deployment methods.
YOLOv8
This project has a YOLOv8 model checkpoint available for inference with Roboflow Deploy. YOLOv8 is a new state-of-the-art real-time object detection model.
Cite This Project
If you use this dataset in a research paper, please cite it using the following BibTeX:
@misc{
traffic-signs-detection-europe_dataset,
title = { Traffic Signs Detection Europe Dataset },
type = { Open Source Dataset },
author = { Radu Oprea },
howpublished = { \url{ https://universe.roboflow.com/radu-oprea-r4xnm/traffic-signs-detection-europe } },
url = { https://universe.roboflow.com/radu-oprea-r4xnm/traffic-signs-detection-europe },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2024 },
month = { feb },
note = { visited on 2024-06-18 },
}
Connect Your Model With Program Logic
Find utilities and guides to help you start using the Traffic Signs Detection Europe project in your project.