Image Augmentation Tool

This Python script helps you augment a small dataset of images to create a larger dataset for training machine learning models. It uses the powerful albumentations library to apply various image transformations to enhance your dataset.

Features

• 🖼️ Multiple Augmentation Techniques: Applies various transformations including:

  • Horizontal flips
  • Random 90° rotations
  • Random rotation (±30°)
  • Brightness and contrast adjustments
  • RGB color shifts
  • Blur effects
  • Gamma corrections
  • Affine transformations (shifting, scaling, rotating)
  • And more!

• 📁 Automatic Folder Handling: Creates output directories if they don't exist

• 🔄 Batch Processing: Processes multiple images in one go

• 🖼️ Multiple Image Formats: Supports .jpg, .jpeg, .png (case insensitive)

• 🛠️ Customizable: Easy to modify and extend for different augmentation needs

Prerequisites

• Python 3.6+
• Required Python packages (install using pip install -r requirements.txt):
  • albumentations
  • opencv-python

Installation

1. Clone this repository:

   git clone [email protected]:Nazmul7989/python-image-augmentation.git
   cd python-image-augmentation

2. Install the required packages:

   pip install -r requirements.txt

Usage

1. Place your input images in the ./images/input/ directory
2. Run the script:

   python main.py

3. Find the augmented images in the ./dataset/augmented/ directory

Configuration

You can modify these settings in main.py:

• input_folder: Path to your input images (default: './images/input')
• output_folder: Where augmented images will be saved (default: './dataset/augmented')
• images_to_generate_per_image: Number of augmented versions to create per image (default: 10)

Folder Structure

.
├── dataset/                 # Output folder for augmented images
│   └── augmented/           # Augmented images will be saved here
├── images/                  # Input folder for original images
│   └── input/               # Place your original images here
├── main.py                 # Main augmentation script
├── requirements.txt        # Python dependencies
└── README.md              # This file

Example

For each input image (e.g., image1.jpg), the script will generate multiple augmented versions (e.g., image1_aug_0.jpg, image1_aug_1.jpg, etc.) in the output folder.

License

This project is licensed under the terms of the MIT license. See the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Acknowledgments

• Built with Albumentations for fast and flexible image augmentations
• Uses OpenCV for image processing

---

Note: This project is designed for general-purpose image augmentation and can be used for various computer vision tasks including but not limited to medical imaging, object detection, and classification tasks.