nprapps / barr-redaction-pixels

Some scripts for working with the image data from the redacted Mueller report

Some scripts for working with the image data from the redacted Mueller report

Prerequisite: You should have Tesseract, Imagemagick, and the Poppler tools installed. Put the report into the directory as report.pdf.

1_print.sh

Creates a directory named pages and extracts images of each page in the PDF into an individually numbered file.

2_measure.sh

Measures the number of colored pixels in each page, and dumps the results into a folder named histograms. Isn't actually useful, since only the text in each redaction bar is colored, not the bar itself.

3_parse.js

Takes the output of 2_measure.sh and generates a CSV with the count of colored pixels on each page.

4_ocr.sh

Requires extracted pages from step 1. Runs the report pages through Tesseract and generates an OCR text file with the report contents.

5_montage.sh

Requires extracted pages from step 1. Generates tiled images in a montages folder consisting of the pages for each major section of the report.