FFT-Based Image Compression

This educational repository provides a Python implementation of image compression using the Fast Fourier Transform (FFT). The code utilizes the Discrete Fourier Transform (DFT) to transform image data into the frequency domain, where less significant frequencies can be discarded to achieve compression.

Results:

PSNR: 129.52 with 12% storage reduction

Usage

Prerequisites

• Python 3.6 or higher
• NumPy
• Pillow (PIL Fork)

Installation

1. Clone the repository:

   git clone https://github.com/CodeKnight314/FFT-Based-Image-Compression.git

2. Install the required packages:

   pip install -r FFT-Based-Image-Compression/requirement.txt

Running the Code

python main.py --img_dir <path_to_image_directory> --threshold <threshold_value> --output_path <output_directory>