Blurring your sensitive information is not safe


It’s becoming much easier to crack internet privacy measures, especially blurred or pixelated images. Those methods make it tough for people to see sensitive information such as obscured license plate numbers or censored faces, but researchers from University of Texas at Austin and Cornell University say that the practice is wildly insecure in the age of machine learning.

Using simple deep learning tools,  the three-person team was able identify obfuscated faces and numbers with alarming accuracy. On an industry standard dataset where humans had 0.19% chance of identifying a face, the algorithm had 71% accuracy (or 83% if allowed to guess five times). The algorithm doesn’t produce a deblurred image—it simply identifies what it sees in the obscured photo, based on information it already knows. The approach works with blurred and pixelated images, as well as P3, a type of JPEG encryption pitched as a secure way to hide information.

Specialized tools for seeing through blur and pixelation have been popping up throughout this year, like the Max Planck Institute’s work on identifying people in blurred Facebook photos. What distinguishes the UT and Cornell research is its simplicity. The attack uses Torch (an open-source deep learning library), Torch templates for neural networks, and standard open-source data.

Even photos obscured by YouTube’s blur feature (center, right) can be recognized.

There are many tools used for recovering passwords from pixelized screenshots. One of such tools is depix. The author of this tool stated:

"Pixelization is used in many areas to obfuscate information in images. I've seen companies pixelize passwords in internal documents. No tools were available for recovering a password from such an image, so I created one. This article covers the algorithm and similar research on depixelization."

The solution is simple: Don't blur your images! Instead, just color over them.


Alvosec is Block Producer for Proton

Download wallet and earn daily staking rewards.

vote for us

Ready for Action?

Don’t hesitate to contact us if you need more information.
Let's Go!
BTC: bc1qnn4zfqqtexl4fkjk2vz6tk74sn92x326wwn0ph

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram