Skip to content

README.md

Latest commit

 

History

History
144 lines (93 loc) · 3.34 KB

README.md

File metadata and controls

144 lines (93 loc) · 3.34 KB

🚘 Automatic License Plate Recognition (ALPR) using Docker

This is a complete AI-powered Automatic License Plate Recognition (ALPR) system built with Python and OpenCV, packaged with Docker for easy deployment. It performs real-time or image-based number plate detection and character recognition using deep learning and OCR. Ideal for smart surveillance, parking systems, toll booths, or traffic enforcement systems.

🔧 Tech Stack

📦 Frameworks & Libraries

  • OpenCV – Image processing and contour detection
  • Pytesseract – Optical Character Recognition (OCR)
  • NumPy – Numerical operations on arrays
  • Matplotlib – Visualizing intermediate steps (optional)
  • Flask (optional) – For building a RESTful API (if using web frontend)

🐳 Tools & Platforms

  • Docker – Containerized deployment
  • Tesseract OCR – External dependency for text recognition
  • Git & GitHub – Version control and collaboration

🖥️ Frontend (Optional)

  • HTML/CSS/JavaScript – Simple UI for uploading images (if web-based)
  • Streamlit – For building an interactive Python dashboard (optional)

🔙 Backend

  • Python – Main programming language for image analysis and OCR
  • Flask – If building as an API backend
  • Docker – To isolate and deploy the app easily

📊 Data Collection Process

📸 Input Data

  • Images of vehicles collected from:

    • Google Images
    • Kaggle datasets
    • Real-time camera footage
    • Custom dataset manually labeled for fine-tuning

📂 Preprocessing Steps:

  • Convert image to grayscale
  • Apply noise reduction
  • Use edge detection (Canny)
  • Find contours and extract number plate region
  • Resize and prepare for OCR

⚙️ Setup & Usage

1️⃣ Clone the Repository

git clone https://github.com/your-username/Automatic-License-Plate-Recognition-Docker.git
cd Automatic-License-Plate-Recognition-Docker

2️⃣ Build and Run with Docker

docker build -t alpr-app .
docker run -it --rm -v ${PWD}/input:/app/input alpr-app

Make sure to place your vehicle images inside the input/ folder before running.

🧠 How It Works

  1. Image Preprocessing

    • Convert to grayscale
    • Edge detection using Canny

  2. License Plate Detection

    • Contour analysis to find plate-like regions

  3. OCR Recognition

    • Tesseract extracts text from the detected plate area

📁 Folder Structure

Automatic-License-Plate-Recognition-Docker/
├── Dockerfile
├── app.py
├── input/            # Folder to store input images
├── output/           # Detected plates and extracted text
├── requirements.txt
└── README.md

🔧 Requirements (for local run without Docker)

  • Python 3.x
  • OpenCV
  • Pytesseract
  • Tesseract OCR (installed on system)

Install dependencies:

pip install -r requirements.txt

📷 Example

Input Image: input_car.jpg

Detected Plate Output: TN01AB1234

(Optional: Add image examples in the repo's examples/ folder.)

🛡️ Use Cases

  • Smart Parking Management Systems
  • Toll Booth Automation
  • Traffic Rule Enforcement
  • Access Control in Secure Zones
  • RTO Vehicle Verification

📝 License

This project is open-source and available under the MIT License.