Linux tip, Fedora tip / howto: convert multiple images to a PDF

 
Note that these tips are mostly outdated


back to notes and tips index

Try my online puzzle page with Calcudoku, Killer Sudoku and online Sudoku.

convert multiple images to a PDF

Let's assume you have scanned 10 pages from a book to files page1.png, page2.png, ..., page10.png. We'll use ImageMagick's convert for the conversion.
  • to reduce the final filesize, you can first reduce the color map size of the files, for example:
    convert page1.png -colors 32 page1b.png
    (increase/decrease the number 32 depending on how many different colors you'd like to see in the final PDF)
  • if you're running tcsh as your shell, you can do all pages in one go by:
    foreach f (1 2 3 4 5 6 7 8 9 10)
    convert page$f.png -colors 32 page${f}b.png
    end
  • if you're running bash, you could do:
    for f in $(seq 10)
    do
    convert page$f.png -colors 32 page${f}b.png
    done
  • convert the resulting images into a single PDF:
    convert page*b.png mydoc.pdf
If you get an incorrect page count (recently I saw this when converting two PNG files to a PDF, it gave me page count of 1), you can try the following:
  • convert page*.png mydoc.ps
  • ps2pdf13 mydoc.ps
  • if stuff from a scanned A4 page does not show up correctly, you can try:
    ps2pdf13 -sPAPERSIZE=a4 mydoc.ps

(so you first convert to PostScript. ps2pdf13 is part of the ghostscript package)

A helpful note from Quique: you can also use convert's option -quality 100 to improve the resulting PDF.


← back to notes and tips index
Please do not copy the text of this tip (© Patrick Min) to your web site.