The Road Sign Recognition project is a real-time detection system designed to recognize road signs across 43 different classes. The project leverages the YOLOv5 model, which is trained on the GTSRB - German Traffic Sign Recognition Benchmark dataset.
This system can be used to improve road safety and assist autonomous driving by identifying and interpreting road signs in real-time through a camera feed.
The project folder road-sign-recognition includes the following subdirectories:
model: Contains the trained YOLOv5 model weights (best.ptandlast.pt).src: Contains the source code for running the project.
- Real-time detection and classification of road signs.
- Recognizes 43 distinct classes of road signs.
- Easy-to-use and customizable codebase.
The model is trained on the GTSRB - German Traffic Sign Recognition Benchmark dataset. This dataset contains over 50,000 images of road signs across 43 different classes, providing a robust training foundation for road sign recognition tasks.
Ensure you have Python installed on your system. The dependencies for the project are listed in the requirements.txt file.
-
Open a terminal or command prompt.
-
Navigate to the project directory:
cd road-sign-recognition -
Install the dependencies using the following command:
pip install -r requirements.txt
-
Ensure you have a webcam or camera feed available.
-
Run the main script located in the
srcfolder to start the real-time detection:python src/main.py -
The system will start detecting and classifying road signs in real-time. Press
qto quit the application.
This project is published as a research paper in the Journal of Emerging Technologies and Innovative Research (JETIR):
Title: Road Sign Recognition using YOLO & GTSRB
Journal: JETIR (Volume 12, Issue 5, May 2025)
Published Paper ID: JETIR2505760
Published Paper Link: JETIR2505760 - Road Sign Recognition using YOLO & GTSRB
- YOLOv5 for object detection.
- GTSRB Dataset for providing the benchmark dataset for training.