Image Recognition (Flask + OpenCV + SQLite)

A simple Flask application for image-based item recognition, leveraging OpenCV's ORB feature descriptor and a SQLite database for storing items and their associated images.

🚀 Features

Item Management

• CRUD Operations: Manage items with Create, Read, Update, and Delete functionalities.
• Multiple Images: Associate multiple images per item.
• Reference URL: Link each item to a reference URL.

Image Processing

• Blurriness Check: Ensure image clarity using Laplacian variance.
• ORB Descriptors: Utilize ORB feature descriptors for reliable feature matching.

Recognition System

• Descriptor Comparison: Compare new image descriptors against stored items.
• Threshold-Based Classification:
  • High Confidence: Confident match found.
  • Borderline: Prompt for more images.
  • No Match: Notify the user of no match.

User Interface

• Image Capture: Capture images via webcam or upload from desktop/mobile.

Authentication

• Single-User Login: Secure login with hashed passwords (bcrypt).
• Session Management: Session-based authentication with configurable lockout after failed attempts.

🛠️ Tech Stack

• Backend: Python (Flask)
• Image Processing: OpenCV
• Database: SQLite
• Frontend: JavaScript, HTML/CSS, Jinja2

📂 Table of Contents

• Features
• Tech Stack
• Getting Started
  • Clone the Repo
  • Install Dependencies
  • Environment Variables
  • Run the Application
• How It Works
  • Item Addition Workflow
  • Recognition Workflow
• Project Structure
• Contributing
• License
• Author

🏁 Getting Started

Clone the Repo

git clone https://github.com/Charisn/Image_recognition.git
cd Image_recognition

Install Dependencies

Ensure Python 3.8+ and pip are installed, then run:

pip install -r requirements.txt

If requirements.txt is missing, manually install key modules: flask, python-dotenv, opencv-python, numpy, bcrypt, etc.

Environment Variables

Create a .env file in the project root:

SECRET_KEY=some-random-key
UPLOAD_FOLDER=static/uploads
MAX_CONTENT_LENGTH_MB=10
ALLOWED_USERNAME=defaultuser
ALLOWED_PASSWORD=defaultpass
MAX_LOGIN_TRIES=5
LOCKOUT_MINUTES=15
DB_PATH=database/items.db
PORT=443
DEBUG=1

Note:

• PORT=443 implies HTTPS. For HTTP (e.g., port 5000), adjust PORT accordingly.
• On first run, the app auto-creates the database and uploads folders.

Run the Application

python app.py

Access the app at https://localhost:443 (or your configured host/port).

🔎 How It Works

Item Addition Workflow

1. Login: Go to /login and enter your credentials.
2. Add Item: Navigate to /add and submit the item's URL.
3. Capture Images: Provide 3 clear images:
   • Blurriness Check: Ensures image clarity using Laplacian variance.
   • ORB Descriptors: Generated and stored for feature matching.
4. Completion: After successful uploads, the item is ready for recognition.

Recognition Workflow

1. Navigate to /view.
2. Capture/Upload an Image.
3. Processing:
   • Blurriness Check: Ensures image clarity.
   • ORB Descriptor Computation: Extract features from the image.
   • Descriptor Comparison: Compare against stored items.
4. Results:
   • High Confidence: Redirect to the item's URL.
   • Borderline: Prompt for more images.
   • No Match: Notify the user of no match.

📁 Project Structure

Image_recognition/
├── app.py                # Main Flask application
├── scripts.js            # Client-side camera handling
├── templates/            # Jinja2 templates
│   ├── base.html
│   ├── index.html
│   ├── login.html
│   ├── add.html
│   ├── capture_images.html
│   ├── view.html
├── static/
│   └── uploads/          # Stored captured images
├── database/
│   └── items.db          # SQLite database (auto-created)
├── .env                  # Environment config
├── requirements.txt      # Project dependencies
└── README.md             # Project documentation

🤝 Contributing

1. Fork the repository

2. Create a feature branch

   git checkout -b feature/new-stuff

3. Commit changes

   git commit -m 'Add new feature'

4. Push to your branch

   git push origin feature/new-stuff

5. Open a Pull Request

📜 License

This project is open-source. Feel free to modify and use it as needed. Check the LICENSE file for more details.

👤 Author

Charisn

Senior Software Engineer passionate about solving problems and exploring new technologies.

Questions or ideas? Open an issue or submit a pull request!