web4355/road-damage-detection

Road Damage Detection Web App

A modern web application for detecting road damage (potholes, cracks, etc.) using YOLOv8 and Flask.

Features

• —Modern UI: Clean, responsive interface with drag-and-drop support.
• —Fast Detection: Uses YOLOv8 for accurate, real-time object detection.
• —Render Ready: Configured for easy deployment to Render (Free Tier compatible).

Tech Stack

• —Backend: Python, Flask, OpenCV, Ultralytics YOLOv8
• —Frontend: HTML5, CSS3, JavaScript (Vanilla)
• —Deployment: Gunicorn (for production)

Running Locally

1. 1.Install Dependencies:

   pip install -r requirements.txt

1. 1.Run the Application:

   python app.py

The app will start at http://localhost:5000.

1. 1.Usage:
2. 2.Open your browser and go to http://localhost:5000.
3. 3.Upload an image (drag & drop or click to browse).
4. 4.View the detected damage and download the result.

Deploying to Render

1. 1.Push this code to a GitHub repository.
2. 2.Log in to Render.
3. 3.Create a new Web Service.
4. 4.Connect your GitHub repository.
5. 5.Render will auto-detect the configuration:
6. 6.Runtime: Python 3
7. 7.Build Command: pip install -r requirements.txt
8. 8.Start Command: gunicorn app:app
9. 9.Click Deploy Web Service.

Note: This app is optimized for Render's Free Tier (512MB RAM). The YOLOv8 model is loaded globally to save memory per request.