Text-to-Bitmap

Problem this Solves

Journalists, activists, and others who need to pass secret messages to prevent interception and decoding by third parties can use Text-to-bitmap to create images that look like random static. The images created are small and can often be hidden in plain sight as part of a larger image (steganography.)

How this project solves this problem

Text-to-Bitmap turns regular text into a scrambled-looking image that can only be decoded with the correct password. Decoding with the incorrect password gives garbage data or false positives.

How to use the tool

Text-to-Bitmap has an API and a command-line tool, both of which can encode text and decode images.

For a demo of the API (no programming necessary), see http://shannonturner.github.io/text-to-bitmap/api.html

For the command line tool, run python uni_text_to_bitmap.py at the terminal to see the full usage details.

How the tool works

Text-to-Bitmap is primarily used to encode a message into a small scrambled-looking image. Text-to-Bitmap can encode messages in any language.

Alice will encode her secret message and send the scrambled image to Bob. They can use a pre-agreed-upon password, or Text-to-Bitmap can automatically create a secure new password for each image.

Either way, when Bob has both the scrambled image and password from Alice, he can use Text-to-Bitmap to decode the image and receive the message.

If Eve intercepts the image, she might attempt to crack the code -- but will most likely be frustrated; each image created by the API has 7,132,766,847,673,865,645,550,849,603,057,000,000 possible passwords. Attempting to decode the image with an incorrect password will either fail outright or worse for Eve, give a false positive result.

Beyond simple messages, Text-to-Bitmap can also encode photos, videos, documents, or any other type of file and turn it into a scrambled image.

Text-to-Bitmap Code breaker

Included with Text-to-Bitmap is a command-line tool to test the strength of the encoded images by systematically trying to break the code.

Using the tool, even running 100 trillion attempts per second would take 22,602,374,222,608,400 years to crack the code.

If you'd like to try to crack the code, see http://shannonturner.github.io/text-to-bitmap/challenges.html for challenges.

Text-to-Smiley (in active development, NOT PRODUCTION READY)

Text-to-Smiley hides a short message in an existing small image of an emoticon by altering the colors imperceptibly. Because the emoticons look identical to the original identicons, Text-to-Smiley's strength is in hiding messages in plain sight (steganography).

Text-to-Smiley is limited to encoding very short messages.

Text-to-Smiley is in the early stages of development and should not yet be used for protecting secret messages.

License

Text-to-Bitmap is free software available under the MIT license; see LICENSE for full details.