Iranian License Plate Recognition

Overview

The Iranian License Plate Recognition project uses machine learning to identify and read license plate numbers from images of Iranian vehicles. It employs computer vision techniques and a logistic regression model to process license plate images and predict the individual digits and characters specific to Iranian license plates.

Features

• Image Processing: Resizes and converts Iranian license plate images to grayscale.
• Data Preparation: Flattens image data and prepares it for machine learning.
• Model Training: Trains a logistic regression model on a dataset of Iranian license plate digits and characters.
• Plate Segmentation: Segments individual characters from a full Iranian license plate image.
• Digit Recognition: Predicts digits and characters using the trained model.
• Visualization: Displays processing steps and results using matplotlib.

Getting Started

To run this project locally, follow these steps:

1. Clone the Repository:

   git clone https://github.com/TahaBakhtari/plate-recognition.git

2. Navigate to the Project Directory:

   cd plate-recognition

3. Install Dependencies:

   pip install numpy opencv-python matplotlib scikit-learn

4. Run the Jupyter Notebook:

   jupyter notebook plate_recognition.ipynb

Dataset

The project uses a custom dataset of Iranian license plate digit and character images stored in numbered folders (1-9). Each folder contains multiple images of the corresponding digit or character.

Key Components

• Image preprocessing and resizing for Iranian license plates
• Feature extraction through image flattening
• Logistic regression model for digit and character classification
• License plate segmentation algorithm tailored for Iranian plates
• Custom prediction function for individual digits and characters

Results

The model achieves an accuracy of approximately 91.15% on the test set. It can successfully segment and recognize digits and characters from full Iranian license plate images.

Iranian License Plate Format

The project is designed to work with the standard Iranian license plate format:

• Two digits
• One letter (in Persian)
• Three digits
• Two digits (usually representing the region code)

Example: 12-365|11 (where | separates the region code)

Future Improvements

• Implement more advanced deep learning models (e.g., CNNs)
• Enhance plate segmentation for varied lighting conditions
• Expand the dataset for improved accuracy on Iranian plates
• Add real-time recognition capabilities for traffic monitoring
• Incorporate Persian character recognition for the letter in the plate

Note:

The number plate photo folders cannot be uploaded on Gainhub due to the large number of them. Please extract the plates_file.zip file and put folders 1 to 9 in the root of the project itself

Note:

This model only learns and predicts numbers and does not have the ability to predict and recognize letters yet !

License

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