Is there a way to flatten a .pdf image from the command line?

Question

In GIMP, I can import a PDF, and use the GUI to flatten it (if it was made with many layers) by selecting Flatten Image in the Image dropdown menu. I can then export the PDF with a new filename.

I would like to automate this. Is there some way to do it via the terminal?

install imagemagick and read man mogrify-im6 , montage-im6 , display-im6 , stream-im6 , identify-im6 , import-im6 , conjure-im6, composite-im6 , convert-im6 , animate-im6 and compare-im6 . — PersianGulf, Oct 19 '14 at 00:29

slm · Accepted Answer · 2020-06-03T00:18:17.767

48

I found these 2 method via Google, in this thread titled: Re: Flattening PDF Files at the UNIX Command Line.

Method #1 - using Imagemagick's convert:

$ convert -density 300 orig.pdf flattened.pdf

NOTE: The quality is reported to be so so with this approach.

Method #2 - Using pdf2ps -> ps2pdf:

$ pdf2ps orig.pdf - | ps2pdf - flattened.pdf

NOTE: This method is reported to retain the image quality.

edited Jun 03 '20 at 00:18

answered Oct 19 '14 at 00:42

slm

363,520
117
767
871

2

Better quality than GIMP on the second -- thanks! – generic_user Oct 19 '14 at 00:48
@ACD - that's good to know. Glad it solved your Q. – slm Oct 19 '14 at 00:49
5

Method 1 generated a very fuzzy image, while method 2 worked perfectly. – Severyn Kozak Jun 20 '15 at 01:06
4

Unfortunately, Method #2 does not rasterize the image, so if you are trying to block out sensitive portions, a user could still open the document and remove layers (with something like Inkscape). You can, however, modify the resolution of Method #1: `$ convert -density 150 {original,flattened}.pdf` If you need to preserve disk space, you can use `-type Grayscale` or `-monochrome` or things of that sort. – eacousineau Oct 08 '16 at 21:28
4

Both methods will rasterize the pdf, although the method #2 does it at a much higher resolution. So none of these methods are satisfying. – Ant Jan 29 '17 at 16:47
2

Regarding Method #1, also see https://stackoverflow.com/questions/6605006/convert-pdf-to-image-with-high-resolution – equaeghe Feb 20 '17 at 22:21
1

I'm using this for resizing notes. Method 2 took ages to convert 12 pages from 2 MB to 70 MB (pdf->ps) and back (ps -> pdf). Ghostscript did it faster. – Just_Alex Apr 01 '20 at 04:24
When I run the `convert` command I get an error. My PDF is only text, no images. error: ```convert: no decode delegate for this image format `' @ error/constitute.c/ReadImage/562. convert: no images defined `no_protect_raster.pdf' @ error/convert.c/ConvertImageCommand/3285.``` – carter Sep 01 '20 at 01:39
1

The second method increase the size of my PDF more than 10x, and lost text highlighting annotation created with Evince. – Alexey Nov 08 '20 at 09:51

score 23 · Answer 2 · edited May 11 '23 at 12:21

23

Ghostscript (gs) worked better than pdf2ps and convert for me. Quality was hardly degraded and file size is small.

gs -dSAFER -dBATCH -dNOPAUSE -dNOCACHE -sDEVICE=pdfwrite \
-sColorConversionStrategy=/LeaveColorUnchanged  \
-dAutoFilterColorImages=true \
-dAutoFilterGrayImages=true \
-dDownsampleMonoImages=true \
-dDownsampleGrayImages=true \
-dDownsampleColorImages=true \
-sOutputFile=document_flat.pdf document_original.pdf

Found here. Note the comment about removing the / before LeaveColorUnchanged.

edited May 11 '23 at 12:21

AdminBee

21,637
21
47
71

answered Oct 06 '15 at 18:06

Stan Bondi

331
2
3

1

This method works great, the pdf is not rasterized and text is conserved as text – Ant Jan 29 '17 at 16:50
5

I tried using this to remove a transparency layer via flattening, but it did not work for me. – mikemtnbikes Nov 06 '18 at 21:13
It changed a font for me but otherwise much better resolution than the other proposed methods. – ste Aug 16 '19 at 09:23
Is there a way to do this with batch files in place? – Just_Alex Apr 01 '20 at 04:40
3

This did not work for me, I get "Unrecoverable error: rangecheck in .putdeviceprops". The ps2pdf solution above did work. – sebastian Mar 22 '21 at 06:46
1

Got also "Unrecoverable error: rangecheck in .putdeviceprops". The ghostscript-answer from user107769 did work. – andreas1724 Jun 20 '21 at 20:26
To fix the `Unrecoverable error: rangecheck in .putdeviceprops` remove the forward slash (`/`) before LeaveColorUnchanged. – Timofey Drozhzhin Oct 29 '22 at 10:09
How would we do this if the image is greyscale? (ie. no need for color.) – Diagon May 11 '23 at 00:22
@Diagon `-sColorConversionStrategy=Gray` (haven't tried it) – Stan Bondi May 23 '23 at 05:01

score 12 · Answer 3 · answered Sep 16 '20 at 21:48

12

ghostscript changed their default for flattening annotations in late 2016.

To simply flatten a PDF now, you need -dPreserveAnnots=false

So a simple command line is now

gs -dSAFER -dBATCH -dNOPAUSE -dNOCACHE -sDEVICE=pdfwrite \
-dPreserveAnnots=false \
-sOutputFile=document_flat.pdf document_original.pdf

answered Sep 16 '20 at 21:48

Lalo

121
1
3

1

Thanks, worked well for me. This command seems to give the same output as Stan Bondi's answer while being simpler. Both were much quicker and gave a smaller file than using `convert`. – Jaydin Nov 19 '20 at 00:16
Ya, ditto. Worked for me too. – Diagon May 11 '23 at 22:03

score 3 · Answer 4 · answered Apr 10 '17 at 17:38

3

Although convert will keep the same file size I've found it to be slow.

The pdf2ps ps2pdf method is faster but I noticed for me it was increasing the file size.

pdftk is nice because it is not only fast but also retains a similar file size.

This is what I use to bulk flatten a directory.

    function pdfflatten () {
        pdftk "$1" output "$2" flatten
    }
    export pdfflatten
    alias pdfflattenDIR='mkdir flattenedPDFs; for i in `seq $(ls *.pdf | wc -l)`; do a=`ls *.pdf | head -$i | tail -1`; pdfflatten "$a" flattenedPDFs/"$a"; done'

answered Apr 10 '17 at 17:38

Jgar

39
1

1

pdftk flatten does _not_ change the images within a PDF. From the docs it "merges an input PDF’s interactive form fields (and their data) with the PDF’s pages" – Aidan Kane Jul 31 '18 at 07:54
3

To be clear, this approach doesn't flatten transparent layers (as indicated above). – mikemtnbikes Nov 06 '18 at 21:15
This works awesome – Dan Ortega Dec 06 '21 at 01:09

score 0 · Answer 5 · answered Nov 23 '20 at 04:00

I frequently need to flatten PDFs or make them image-only. I wanted to do this quickly from within Nautilus instead of going to a terminal, so I created two Nautilus scripts based on the previous answers. I am sharing the instructions in case they are useful to others.

Create script to flatten PDFs:

Go to the Nautilus script folder ~/.local/share/nautilus/scripts/
Create a new shell script file (eg FlattenPDF.sh)
Make this file executable (eg Natiuls Properties → Permissions → Allow executing file as program)
Insert below text into the file:

#!/bin/bash

# Flattend PDFs    

IFS='
'
for file in $NAUTILUS_SCRIPT_SELECTED_FILE_PATHS
do
  if [ -f "$file" ]; then
    base=${file%.*}
    ext=${file##*.}
    newname=${base}_flat.pdf
    gs -dSAFER -dBATCH -dNOPAUSE -dNOCACHE -sDEVICE=pdfwrite -dPreserveAnnots=false -sOutputFile=$newname $file
  fi
done

Create script to create image-only (text not searchable or selectable) PDFs:

Go to the Nautilus script folder ~/.local/share/nautilus/scripts/
Create a new shell script file (eg ImageOnlyPDF.sh)
Make this file executable (eg Natiuls Properties → Permissions → Allow executing file as program)
Insert below text into the file:

#!/bin/bash

# Image-only PDFs    

IFS='
'
for file in $NAUTILUS_SCRIPT_SELECTED_FILE_PATHS
do
  if [ -f "$file" ]; then
    base=${file%.*}
    ext=${file##*.}
    newname=${base}_img.pdf
    gconvert -density 200 $file $newname
  fi
done

Restart nautilus (eg type nautilus -q to quit Nautilus and then open Nautilus from the Dash again). If you right-click on a file, you should see the Scripts menu with the new scripts.

Is there a way to flatten a .pdf image from the command line?

5 Answers5

Linked