Hybrid Vision Real Time Traffic Sign Detection Classification YOLOv8 Res Net50

Abstract

1. Introduction

HybridVision is a real-time traffic sign recognition system built with a hybrid deep learning pipeline for both detection and fine-grained classification, specially focused on Bangladeshi road environments. It combines a YOLO detector to localize signs (bounding boxes) and a ResNet50 classifier to identify the exact traffic sign category from cropped sign regions. A confidence-based fusion strategy is used to produce stable final predictions.

Custom Dataset: 9,133 images (Bangladeshi traffic signs) Total Classes: 29 sign categories Trained on: Kaggle GPU (P100) Results (reported): mAP 100% / Precision 100% / Recall 100% / Accuracy 99% Deployed as Web App: Streamlit (Image / Video / Live Camera)

2. Objective

HybridVision is a real-time traffic sign recognition system built with a hybrid deep learning pipeline for both detection and classification. I created a custom Bangladeshi traffic sign dataset by collecting and annotating 9,133 images, organized into 29 traffic sign categories using Roboflow.

This project demonstrates how Jupyter Notebook can be applied to a real-world AI & Machine Learning problem.

3. Key Features / Modules

• Real-time traffic sign detection + classification
• Built a custom Bangladeshi dataset (9,133 images)
• 29 sign categories
• Experimented with YOLOv5, YOLOv7, YOLOv8, YOLOv11
• ResNet-based classification, finalized ResNet50
• End-to-end ML workflow:
• dataset creation → preprocessing/augmentation → training → evaluation → packaging → deployment
• Streamlit Web UI:
• Image Upload
• Video Upload

4. Technology Stack

5. System Requirements

General requirements for this technology stack — check the README for exact versions.

• Python 3.8 or later with Jupyter Notebook / JupyterLab (or Google Colab)
• pip for dependencies
• Git (to clone the repository)

6. Installation & Setup

git clone https://github.com/masumtanjim7/HybridVision-Real-Time-Traffic-Sign-Detection-Classification-YOLOv8-ResNet50.git
cd HybridVision-Real-Time-Traffic-Sign-Detection-Classification-YOLOv8-ResNet50

1. Detector: YOLOv7 (real-time bounding boxes)
2. Classifier: ResNet50 (cropped sign classification)
3. Fusion: confidence-based final decision

Full setup instructions are in the project README.

7. Future Enhancements

Suggested extensions you can add to make this your own project.

• Deploy the model as a web app with Streamlit, Flask or FastAPI
• Compare against an additional model and report the metric difference
• Add explainability (SHAP / Grad-CAM)

8. Viva / Review Questions

Common questions examiners ask for projects in this domain.

1. What dataset does the project use and how was it pre-processed?
2. Which algorithm / model architecture is used and why was it chosen over alternatives?
3. How are training and testing data split, and how is overfitting avoided?
4. Which evaluation metrics (accuracy, precision, recall, F1) are reported and what do they mean here?
5. How would you deploy this model for real users?

9. Source Code & License

This project is developed by masumtanjim7 and published on GitHub under the GNU General Public License v3.0. Please follow the license terms and credit the original author when you use or modify this code.