Skip to content

A tiny essay on image inside docx anonymization. So far just works with faces (not all faces yet)

Notifications You must be signed in to change notification settings

casanovajose/blurea_tutti

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BlureaTutti

A tiny essay on image anonymization. So far just works with faces (not all faces yet)

What this script do

  • Extracts images from docx files
  • Extracts rectangles of faces using OpenCV Cascade classifier from extracted images
  • Applies Gaussian Blur over the ROIs
  • Saves anonymized images in a new folder

Usage

I'm using jjanzic/docker-python3-opencv to easy install of OpenCV in Linux.

  • docker pull jjanzic/docker-python3-opencv
  • clone this repo
  • load the docs folder with your docs
  • docker run --rm --name opencv -v $(pwd):/tmp -it jjanzic/docker-python3-opencv /bin/bash
  • cd tmp
  • pip install -r requirements.txt
  • python test.py
  • see the results in anonymized folder

Example

The input document

Original Doc

The anonymized output

Extracted/Anonymized

Extracted/Anonymized

TODO

  • Clean images folder after anonym....
  • Add other clasiffiers (licence plates, etc.)
  • Improve docker integration
  • A lot of more things :)

About

A tiny essay on image inside docx anonymization. So far just works with faces (not all faces yet)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages