Skip to content

Latest commit

 

History

38 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AI Road Damage Detection System

Deep Learning | YOLOv11 | Streamlit

An AI-powered Road Damage Detection System that automatically detects and localizes different types of road damage from images using the RDD2022 Dataset and YOLOv11.

Python YOLO PyTorch Streamlit OpenCV


Overview

Road damage inspection is traditionally performed manually, making the process time-consuming and difficult to scale.

This project presents an AI-powered Computer Vision application that automatically detects road damage from images using a trained YOLOv11 object detection model.

The application provides an interactive Streamlit interface where users can upload road images and view detected damage with bounding boxes and confidence scores.


Features

  • Road Damage Detection
  • Crack Detection
  • Pothole Detection
  • Detection of Multiple Road Damage Categories
  • Bounding Box Visualization
  • Confidence Score Analysis
  • Image Upload
  • YOLOv11 Object Detection
  • Interactive Streamlit Dashboard
  • Analytics Dashboard
  • Detection History
  • Annotated Detection Results
  • User-Friendly Interface

Damage Classes

Class Description
D00 Longitudinal Crack
D10 Transverse Crack
D20 Alligator Crack
D40 Pothole

System Workflow

Road Image
     |
     v
Image Upload
     |
     v
Image Preprocessing
     |
     v
YOLOv11 Detection Model
     |
     v
Damage Detection
     |
     v
Bounding Boxes + Confidence Scores
     |
     v
Detection Results

Technology Stack

Programming Language

  • Python

Deep Learning

  • YOLOv11
  • PyTorch
  • Ultralytics

Computer Vision

  • OpenCV
  • Pillow

Data Processing

  • NumPy
  • Pandas

Visualization

  • Matplotlib

Web Application

  • Streamlit

Development Tools

  • Git
  • GitHub
  • VS Code
  • Jupyter Notebook

Project Structure

road-damage-ai-detection/
|
├── frontend/
│   ├── app.py
│   ├── assets/
│   │   └── logo.png
│   └── ...
│
├── notebooks/
│   └── ...
│
├── requirements.txt
├── packages.txt
├── README.md
├── .gitignore
└── LICENSE

Dataset

RDD2022 - Road Damage Detection 2022

The project is based on the Road Damage Detection 2022 (RDD2022) dataset.

The dataset contains annotated road images collected from multiple countries and includes different types of road surface damage.

The model is trained to detect the following road damage categories:

  • D00 - Longitudinal Crack
  • D10 - Transverse Crack
  • D20 - Alligator Crack
  • D40 - Pothole

Installation

Clone Repository

git clone https://github.com/WardaAhad/road-damage-ai-detection.git
cd road-damage-ai-detection

Install Dependencies

pip install -r requirements.txt

Run the Application

streamlit run frontend/app.py

The application will be available at:

http://localhost:8501

Application Sections

Dashboard

Provides an overview of the Road Damage Detection System and project information.

Detect Damage

Upload a road image and use the trained YOLOv11 model to detect road damage.

Analytics

Provides visual insights and statistics related to detection results.

History

Allows users to review previously processed detection results.


Application Screenshots

Dashboard

Add Dashboard Screenshot here.

Detection

Add Detection Screenshot here.

Analytics

Add Analytics Screenshot here.

History

Add History Screenshot here.


Live Demo

Try the deployed Streamlit application:

https://road-damage-ai-detection-zlum5wd8ojrh6toetrnlqb.streamlit.app/


Project Objectives

  • Automate road damage detection using Computer Vision.
  • Apply YOLOv11 object detection to real-world road images.
  • Detect and localize multiple types of road damage.
  • Build an interactive AI application using Streamlit.
  • Visualize detection results using bounding boxes and confidence scores.
  • Gain practical experience in Deep Learning and Computer Vision.

Future Improvements

  • Video Road Damage Detection
  • Live Camera Detection
  • GPS Integration
  • Automated PDF Reports
  • Damage Severity Analysis
  • Interactive Road Damage Maps
  • Mobile Application
  • Cloud-Based Storage
  • Real-Time Monitoring

Developer

Warda Ahad

Bachelor of Artificial Intelligence

The Islamia University of Bahawalpur

GitHub:

https://github.com/WardaAhad


Show Your Support

If you find this project useful, consider giving the repository a star on GitHub.

Your support helps encourage further development and learning.


License

This project is licensed under the MIT License.


AI for Safer Roads

Built with Python, YOLOv11, and Streamlit.

Contributors

Languages