Steg Project

This is an exploration into Steganography using Python. Messages can be hidden within audio or image files.

Supported formats

Many image/audio formats should work as input files. The actual decoding is delegated to the Pillow library for images, and ffmpeg for audio. Any of their input formats should be acceptable.

That being said, only lossless output formats can be utilized. If lossy compression were used, the least significant bits (LSBs) where the message is stored could be squashed.

Supported image output formats:

"bmp", "im", "j2k", "pbm", "pcx", "pgm", "ppm", "png", "tiff", "tif", "webp" 

Supported audio output formats:

"aiff", "ast", "au", "caf", "f32be", "f32le", "f64be", "f64le", "flac", "ircam", "s16be", "s16le", "s24be", "s24le", "s32be", "s32le", "smjpeg", "sox", "u16be", "u16le", "u24be", "u24le", "u32be","u32le", "voc", "w64", "wav", "wv"

Command Line Interface

First, run the setuptools to prepare the executable.

sudo -H pip install --editable .

Reading and writing plain text

Text can be embedded through stdin, and terminated with CTRL+D

steg conceal - tests/images/halle.png shakey_halle.png
This is the text I want to hide in my image.

It can then be read back:

steg reveal shakey_halle.png -
This is the text I want to hide in my image.

Reading and writing files

Files can be handled in a similar fashion. The following embeds all the works of Shakespeare (compressed) into an image:

steg conceal tests/text/shakespeare_all.bz2 tests/images/halle.png shakey_halle.png

It can then be extracted back out:

steg reveal shakey_halle.png shakespeare_extracted.bz2

Fun with pipes

Since the CLI reads from standard input, data can be piped through other utilies which perform other transformations. The following encrypts and compresses the text "this is some text" and conceals it in an image.

echo this is some text | gzip | gpg -c --passphrase my_password | steg conceal - tests/images/halle.png shakey_halle.png

To recover the plain text, execute the following:

steg reveal shakey_halle.png - | gpg -d -q --passphrase my_password | gzip -d

Audio files supported as well

All of the above examples work equally well with audio files as well. Here's the previous example using .wav files

echo this is some text | gzip | gpg -c --passphrase my_password | steg conceal - tests/audio/bach_standard.wav bach_stego.wav

To recover the plain text, execute the following:

steg reveal bach_stego.wav - | gpg -d -q --passphrase my_password | gzip -d

Dev Env:

Ubuntu 16.04 LTS

Setup:

Python Dev Dependencies:
    sudo apt install python3-dev python3-setuptools python-pip
    pip install --upgrade pip

Pillow Imaging Dependencies
    sudo apt install libtiff5-dev libjpeg8-dev zlib1g-dev libfreetype6-dev liblcms2-dev libwebp-dev tcl8.5-dev tk8.5-dev python-tk

Pydub Audio Dependencies
    sudo apt install ffmpeg libavcodec-extra

Project Dependencies:

If unable to setuptools or the requirements.txt for some reason:

pip install Pillow
pip install bitarray
pip install Click
pip install pydub