On the (In)effectiveness of Mosaicing and Blurring as Tools for Document Redaction

By Steven Hill, Zhimin Zhou, Lawrence K. Saul, Hovav Shacham. 2016

79 citations recorded by the source corpus.

Abstract In many online communities, it is the norm to redact names and other sensitive text from posted screenshots. Sometimes solid bars are used; sometimes a blur or other image transform is used. We consider the effectiveness of two popular image transforms - mosaicing (also known as pixelization) and blurring - for redaction of text. Our main finding is that we can use a simple but powerful class of statistical models - so-called hidden Markov models (HMMs) - to recover both short and indefinitely long instances of redacted text. Our approach borrows on the success of HMMs for automatic speech recognition, where they are used to recover sequences of phonemes from utterances of speech. Here we use HMMs in an analogous way to recover sequences of characters from images of redacted text. We evaluate an implementation of our system against multiple typefaces, font sizes, grid sizes, pixel offsets, and levels of noise. We also decode numerous real-world examples of redacted text. We conclude that mosaicing and blurring, despite their widespread usage, are not viable approaches for text redaction.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Vitaly Shmatikov
2. Richard McPherson
3. Ying Zhao
4. Jinjun Chen
5. Reza Shokri
6. Mario Fritz
7. Bernt Schiele

In Signal Processing

1. Richard McPherson
2. Vitaly Shmatikov
3. Ying Zhao
4. Jinjun Chen
5. Reza Shokri
6. Joseph Bonneau

Citations in the corpus

• Defeating Image Obfuscation with Deep Learning (2016, methodology): “The average color of every pixel in each square is computed and the entire square is set to that color [16].”
• A Survey on Differential Privacy for Unstructured Data Content (2022, background): “However, several recent studies pointed out that these traditional methods are not sufficient to guarantee image privacy due to the fact that machine learning models can be adapted to re-identify faces and texts [47].”