If you are unfamiliar with steganography techniques, I have also written a basic overview of the field in Steganography: Hiding Data Inside Data.
This project is on PyPI and can be installed with
pip install stego-lsb
Alternatively, you can install it from this repository directly:
git clone https://github.com/ragibson/Steganography
cd Steganography
python3 setup.py install
After installation, use the stegolsb command in the terminal or import
functions from stego_lsb in your code.
bit_manipulation provides the ability to (quickly) interleave the bytes of a payload directly in the least significant bits of a carrier byte sequence.
Specifically, it contains four primary functions:
# Interleave the bytes of payload into the num_lsb LSBs of carrier.
lsb_interleave_bytes(carrier, payload, num_lsb, truncate=False)
# Deinterleave num_bits bits from the num_lsb LSBs of carrier.
lsb_deinterleave_bytes(carrier, num_bits, num_lsb)
# Runs lsb_interleave_bytes with a List[uint8] carrier.
lsb_interleave_list(carrier, payload, num_lsb)
# Runs lsb_deinterleave_bytes with a List[uint8] carrier.
lsb_deinterleave_list(carrier, num_bits, num_lsb)
Running bit_manipulation.py, calling its test() function directly, or
running stegolsb test should produce output similar to
Testing 1.0 MB payload -> 10.0 MB carrier...
Progress: [################################]
----------------------------------------
| # LSBs | Encode Rate | Decode rate |
| 1 | 60.6 MB/s | 95.9 MB/s |
| 2 | 56.6 MB/s | 52.7 MB/s |
| 3 | 82.5 MB/s | 77.4 MB/s |
| 4 | 112.4 MB/s | 105.9 MB/s |
| 5 | 135.9 MB/s | 129.8 MB/s |
| 6 | 159.9 MB/s | 152.4 MB/s |
| 7 | 181.7 MB/s | 174.6 MB/s |
| 8 | 372.8 MB/s | 1121.8 MB/s |
----------------------------------------
WavSteg uses least significant bit steganography to hide a file in the samples of a .wav file.
For each sample in the audio file, we overwrite the least significant bits with the data from our file.
WavSteg requires Python 3
Run WavSteg with the following command line arguments:
Command Line Arguments:
-h, --hide To hide data in a sound file
-r, --recover To recover data from a sound file
-i, --input TEXT Path to a .wav file
-s, --secret TEXT Path to a file to hide in the sound file
-o, --output TEXT Path to an output file
-n, --lsb-count INTEGER How many LSBs to use [default: 2]
-b, --bytes INTEGER How many bytes to recover from the sound file
--help Show this message and exit.
Example:
$ stegolsb wavsteg -h -i sound.wav -s file.txt -o sound_steg.wav -n 1
# OR
$ stegolsb wavsteg -r -i sound_steg.wav -o output.txt -n 1 -b 1000
Hiding data uses the arguments -h, -i, -s, -o, and -n.
The following command would hide the contents of file.txt into sound.wav and save the result as sound_steg.wav. The command also outputs how many bytes have been used out of a theoretical maximum.
Example:
$ stegolsb wavsteg -h -i sound.wav -s file.txt -o sound_steg.wav -n 2
Using 2 LSBs, we can hide 6551441 bytes
Files read in 0.01s
5589889 bytes hidden in 0.24s
Output wav written in 0.03s
If you attempt to hide too much data, WavSteg will print the minimum number of LSBs required to hide your data.
Recovering data uses the arguments -r, -i, -o, -n, and -b
The following command would recover the hidden data from sound_steg.wav and save it as output.txt. This requires the size in bytes of the hidden data to be accurate or the result may be too short or contain extraneous data.
Example:
$ stegolsb wavsteg -r -i sound_steg.wav -o output.txt -n 2 -b 5589889
Files read in 0.02s
Recovered 5589889 bytes in 0.18s
Written output file in 0.00s
LSBSteg uses least significant bit steganography to hide a file in the color information of an RGB image (.bmp or .png).
For each color channel (e.g., R, G, and B) in each pixel of the image, we overwrite the least significant bits of the color value with the data from our file. In order to make recovering this data easier, we also hide the file size of our input file in the first few color channels of the image.
You need Python 3 and Pillow, a fork of the Python Imaging Library (PIL).
Run LSBSteg with the following command line arguments:
Command Line Arguments:
-h, --hide To hide data in an image file
-r, --recover To recover data from an image file
-a, --analyze Print how much data can be hidden within an image [default: False]
-i, --input TEXT Path to an bitmap (.bmp or .png) image
-s, --secret TEXT Path to a file to hide in the image
-o, --output TEXT Path to an output file
-n, --lsb-count INTEGER How many LSBs to use [default: 2]
-c, --compression INTEGER RANGE
1 (best speed) to 9 (smallest file size) [default: 1]
--help Show this message and exit.
Example:
$ stegolsb steglsb -a -i input_image.png -s input_file.zip -n 2
# OR
$ stegolsb steglsb -h -i input_image.png -s input_file.zip -o steg.png -n 2 -c 1
# OR
$ stegolsb steglsb -r -i steg.png -o output_file.zip -n 2
Before hiding data in an image, it can be useful to see how much data can be hidden. The following command will achieve this, producing output similar to
$ stegolsb steglsb -a -i input_image.png -s input_file.zip -n 2
Image resolution: (2000, 1100, 3)
Using 2 LSBs, we can hide: 1650000 B
Size of input file: 1566763 B
File size tag: 3 B
The following command will hide data in the input image and write the result to the steganographed image, producing output similar to
$ stegolsb steglsb -h -i input_image.png -s input_file.zip -o steg.png -n 2 -c 1
Files read in 0.26s
1566763 bytes hidden in 0.31s
Image overwritten in 0.27s
The following command will recover data from the steganographed image and write the result to the output file, producing output similar to
$ stegolsb steglsb -r -i steg.png -o output_file.zip -n 2
Files read in 0.30s
1566763 bytes recovered in 0.28s
Output file written in 0.00s
StegDetect provides one method for detecting simple steganography in images.
You need Python 3 and Pillow, a fork of the Python Imaging Library (PIL).
Run StegDetect with the following command line arguments:
Command Line Arguments:
-i, --input TEXT Path to an image
-n, --lsb-count INTEGER How many LSBs to display [default: 2]
--help Show this message and exit.
We sum the least significant n bits of the RGB color channels for each pixel and normalize the result to the range 0-255. This value is then applied to each color channel for the pixel. Where n is the number of least significant bits to show, the following command will save the resulting image, appending "_nLSBs" to the file name, and will produce output similar to the following:
$ stegolsb stegdetect -i input_image.png -n 2
Runtime: 0.63s