Road Sign Recognition Project

Overview

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.

Project Structure

The project folder road-sign-recognition includes the following subdirectories:

1. model: Contains the trained YOLOv5 model weights (best.pt and last.pt).
2. src: Contains the source code for running the project.

Features

• Real-time detection and classification of road signs.
• Recognizes 43 distinct classes of road signs.
• Easy-to-use and customizable codebase.

Dataset

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.

Requirements

Ensure you have Python installed on your system. The dependencies for the project are listed in the requirements.txt file.

Steps to Install Dependencies

1. Open a terminal or command prompt.

2. Navigate to the project directory:

   cd road-sign-recognition
   

3. Install the dependencies using the following command:

   pip install -r requirements.txt
   

How to Run the Project

1. Ensure you have a webcam or camera feed available.

2. Run the main script located in the src folder to start the real-time detection:

   python src/main.py
   

3. The system will start detecting and classifying road signs in real-time. Press q to quit the application.

Model Training Results

Acknowledgements

• YOLOv5 for object detection.
• GTSRB Dataset for providing the benchmark dataset for training.