LSB Image Steganography

Python and JS implementations of the Least Significant Bit (LSB) Image Steganography method.

Python

The dependencies for the Python version can be installed with pip install -r requirements.txt.

The implementation is all in the lsb_image_steg.py file; it can be used by importing (e.g.: from lsb_image_steg import LSBImageSteganography) and then instantiating the class with the image where the data will be encoded/ from which it will be decoded (lsb = new LSBImageSteganography('image.png')) and using its encode and decode methods. The parameters are documented in the source code. When encoding new data, the image can be saved with the save method.

The test_steg.py file contains the unit tests for the implementation and can be run with python3 -m unittest test_steg.py.

JS

The main implementation is in pure JS in the lsb_image_steg.js file, from which the loadImage, encode and decode functions can be imported and used. The parameters are documented in the source code. Note that loadImage must be called first, to load the image into a canvas (if running on Node, the node-canvas library is a dependency), and then data can be encoded into/ decoded from the image in the canvas.

Although the functions will execute without errors when using them, images containing an alpha channel are not recommended, as they can impact the performance and generate outputs with visual artifacts or changes.

There's also a web app (available here) with an interface for applying the steganography method, implemented in Svelte. To run this app, first install the dependencies with npm i then use the dev server with npm run dev, or use npm run build to generate the bundle or npm run preview to better test it.