Traffic-Sign-Detection-YOLOv8

🚦 Traffic Sign Detection using YOLOv8

📌 Overview

This repository contains a custom-trained YOLOv8 object detection model for detecting and classifying traffic signs under various road and environmental conditions.

The model was trained using the Ultralytics YOLOv8 framework and can identify 36 different traffic sign classes.

It is intended for research, educational purposes, and computer vision applications such as intelligent transportation systems and autonomous driving.

---

✨ Features

• Detects 36 traffic sign classes
• Built using Ultralytics YOLOv8
• Fine-tuned from a pretrained YOLOv8 model
• Fast inference with high accuracy
• Supports images, videos, and webcam inference

---

🧠 Model Details

---

📊 Performance

The repository includes the following evaluation files:

• Training Results
• Precision Curve
• Recall Curve
• Precision-Recall Curve
• F1 Curve
• Confusion Matrix
• Normalized Confusion Matrix
• Label Distribution

These files are available inside the repository for detailed performance analysis.

---

📂 Repository Contents

best.pt                          Trained model weights
results.png                      Training metrics
results.csv                      Training log
confusion_matrix.png             Confusion Matrix
confusion_matrix_normalized.png  Normalized Confusion Matrix
BoxP_curve.png                   Precision Curve
BoxR_curve.png                   Recall Curve
BoxPR_curve.png                  Precision-Recall Curve
BoxF1_curve.png                  F1 Curve
labels.jpg                       Label Distribution

---

🚀 Quick Start

Install dependencies

pip install ultralytics

Load the model

from ultralytics import YOLO

model = YOLO("best.pt")

Run inference

results = model("image.jpg")

results[0].show()

---

📈 Training

The model was trained using the Ultralytics YOLOv8 framework on a custom traffic sign dataset.

Training configuration and logs are included within this repository.

---

⚠️ Limitations

• Performance depends on image quality.
• Extremely small or heavily occluded traffic signs may reduce detection accuracy.
• Additional training data may improve performance in different geographic regions.

---

📜 License

This project is released under the MIT License.

---

👤 Author

Ankit Kumar Jha

B.Tech (Data Science)

Machine Learning | Computer Vision | Python