This project is an image compression tool implemented in Rust, leveraging the K-Means clustering algorithm to reduce the color palette of an image. This method can significantly decrease the size of the image file while maintaining visual fidelity to a degree determined by the number of clusters (K) used in the compression process.

• CLI for easy use and integration into workflows.
• Customizable number of clusters for flexible compression options.
• Utilizes rayon for parallel processing, improving performance on multi-core systems.
• Input and output path flexibility.

Before you begin, ensure you have met the following requirements:

• Rust and Cargo (Rust's package manager) installed on your machine.
• Basic familiarity with Rust and command-line operations.

Clone the repository to your local machine:

git clone https://github.com/nethriis/image-compressor.git
cd image-compressor

Build the project using Cargo:

cargo build --release

The executable will be located in ./target/release/.

To compress an image, run the tool from the command line, specifying the input image path, the output image path, and optionally, the number of clusters (K). The default value for K is 4 if not specified.

./target/release/image_compressor -i path/to/input.png -o path/to/output.png -k 16

• -i, --input: The path to the input image file.
• -o, --output: The path where the compressed image will be saved.
• -k, --k: (Optional) The number of color clusters to use for compression. A higher number retains more detail but reduces compression.

Contributions to this project are welcome. Please adhere to the following guidelines:

• Fork the repository and create your branch from main.
• Write clear and concise commit messages.
• Ensure any install or build dependencies are removed before the end of the layer when doing a build.
• Update the README.md with details of changes to the interface, this includes new environment variables, exposed ports, useful file locations, and container parameters.

Distributed under the GPL-3.0 license.

Generated with MarkdownView ✨