Skip to content

Usage Guide

This guide explains how to use the AIC-HCMUS Fragment Segmentation application, from setup to making predictions.

1. Prerequisites

  • Docker and Docker Compose installed
  • Modern web browser (for frontend)
  • (Optional) Python 3.10+ and Node.js

2. Running the Application

Using Docker Compose

  1. Clone the repository: 
    git clone https://github.com/magnusdtd/AIC-HCMUS-Fragment-Segmentation.git
cd AIC-HCMUS-Fragment-Segmentation

  2. Generate a self-signed SSL certificate (openssl is required): 

    mkdir -p nginx-ssl
openssl req -x509 -nodes -days 365 -newkey rsa:2048 \
  -keyout nginx-ssl/tls.key \
  -out nginx-ssl/tls.crt \
  -subj "/C=US/ST=State/L=City/O=Organization/OU=Unit/CN=localhost"

  3. Update the GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET in docker-compose.yml if you want to use your credential.

  4. Start all services: 

    docker compose up --build

  5. Access the frontend at http://localhost:443

3. App Features

Register

  • Open the web app.
  • Press the register button to navigate to the register page.
  • Enter your username and password.
  • The page will automatically redirect to the default page.

Manual Register

2. Login

2.1. Manual login

Manual Login

2.2. Google Login

Google Login

3. Upload and Predict on an Image

  • After logging into the app, navigate to the predict tab. Predict Tab
  • Upload an image file (e.g., a fragment image).
  • Adjust the values for real radius, unit, confidence (conf), or Intersection over Union (iou) if needed.
  • Submit to receive segmentation results.

Segmentation Result 1 Segmentation Result 2

4. View Uploaded Images

  • The images tab contains all user-uploaded images. User Images
  • Click on each image to see its details. Image Details
  • Use the view button to see prediction details and the download button to download artifacts.

For further help, see the About page or open an issue on GitHub.