Convert PDF to black and white CCITT Groupe 4 one. Use ImageMagisk and bbe or sed (if available).

README.md

Description

This script needs the following tools :

• ImageMagick, convert (Mac/Linux) or magick (Windows), to convert all .png pages to black and white and compress them with CCITT Groupe 4 compression.
• Use bbe or sed (if present) to add interpolation option to all CCITT streams. This makes the pages more pleasing to the eye (less "pixelated").

Usage

> ./pdf2bw.sh
Usage: pdf2bw.sh file.pdf [options]
Result: file_bw.pdf
Options:
  -r, --resolution <int> : the resolution in dpi [default 200]
  -t, --thershold <0-100> : the black/white thershold in percent [default 77]

pdf2bw.sh

#!/bin/sh
# https://gist.github.com/kpym/dc74bdc78a192add4ca4b3853fb3473d

# Convert PDF to black and white CCITT Groupe 4 using ImageMagisk
# and add blure filter with bbe or sed (if available).

# ============== SET PARAMETERS =====================================

# convert arguments to variables
while [ -n "$1" ]; do
key="$1"
shift # past argument
  case $key in
      -r|--resolution)
        resolution="$1"
        shift # past value
      ;;
      -t|--threshold)
        thershold="$1"
        shift # past value
      ;;
      *.pdf) # the pdf
        base="${key%.pdf}" # The pdf name without the extension .pdf
      ;;
      *) # unknown option
        echo "Error: Unknown parameter $key"
        base=""
  esac
done

# Print the usage message (if needed)
if [ -z "$base" ]; then
    echo "Usage: $0 file.pdf [options]"
    echo "Result: file_bw.pdf"
    echo "Options:"
    echo "  -r, --resolution <int> : the resolution in dpi [default 200]"
    echo "  -t, --thershold <0-100> : the black/white thershold in percent [default 77]"
    exit 1
fi

# Check for ImageMagisk = magick on Windows, convert on Mac/Linux
if command -v magick &> /dev/null
then
  im=magick
  echo "Use magick (probably on Windows)"
else
  if command -v convert &> /dev/null
  then
    im=convert
    echo "Use convert (hope is not Windows)"
  else
    echo "Error: can'n find ImageMagick (magick or convert)."
    exit 1
  fi
fi

# Set the output resolution in dpi
if [ -z "$resolution" ]; then
    resolution=200
fi
echo "Resolution set to ${resolution}dpi."

# Set the black and white thershold
if [ -z "$thershold" ]; then
    thershold=77
fi
echo "Thersnold set to ${thershold}%."

# ===================== CONVERT =====================================

# Do the actual conversion
echo "Converting ${base}.pdf to ccitt black and white pdf."
$im -density $resolution "${base}.pdf" -threshold "${thershold}%" -type bilevel -compress Group4 "${base}_temp.pdf"

# ==================== ADD BLUR =====================================

# Use bbe or sed (if present) to add interpolation option to CCITT strames
if command -v bbe &> /dev/null; then
  echo "Add interpolation (blur) using bbe : ${base}_temp.pdf -> ${base}_bw.pdf"
  bbe -b "/BitsPerComponent 1/:18" -e "A /Interpolate\32true" "${base}_temp.pdf" -o "${base}_bw.pdf"
elif command -v sed &> /dev/null; then
  echo "Add interpolation (blur) using sed : ${base}_temp.pdf -> ${base}_bw.pdf"
  sed -b -e 's;/BitsPerComponent 1;/BitsPerComponent 1/Interpolate true;' "${base}_temp.pdf" > "${base}_bw.pdf"
else
  echo "Can't find bbe nor sed. The images will look very sharp."
  echo "Move: ${base}_temp.pdf -> ${base}_bw.pdf"
  mv "${base}_temp.pdf" "${base}_bw.pdf"
fi

# ====================== CLEAR ======================================

# Clear temporary files
if [ -f "${base}_temp.pdf" ]; then
  echo delete temp pdf
  rm "${base}_temp.pdf"
fi