Duplex Scan Left Blank Pages in Your PDF? Delete Them

The cyber cafe's scanner was set to two-sided. Your twelve-page agreement came back as a twenty-four page PDF: page 1 is the agreement, page 2 is a grey rectangle, page 3 is the agreement again, page 4 is another grey rectangle. All the way down.

The document is fine. It is just twice as long as it should be, twice as heavy, and it looks careless when the other party opens it.

Why the backs are not actually blank

Worth understanding before you go hunting for an automatic "delete blank pages" button, because that button exists in some software and it will let you down here.

The backs of your sheets are not empty white. A duplex scanner captures whatever the sensor sees, and on a real sheet of paper that includes sensor noise, a faint grey cast from the scanner lid, dust specks, and show-through — the ghost of the printing on the other side bleeding faintly through the paper. To a program checking whether a page is blank, a page with 0.4% of its pixels non-white is not blank.

So automatic detection either misses them, or gets aggressive and starts eating pages that contain a single line of text. Neither outcome is what you want on a document with signatures on it.

Typing the numbers yourself takes ninety seconds and it is right the first time.

The walkthrough

1. Load the file into the Page Remover. Next to the filename you get the page count — 24 pages — which is your confirmation that the duplex theory is correct. Twelve sheets, twenty-four sides.

2. Scroll the preview and check the pattern. Do not assume every even page is blank. Assumptions break in three specific places:

• The last sheet, if the agreement ended on an odd side, may have a genuinely blank back — but it may also have the stamp or a witness signature on it.
• If the operator scanned in two goes, the pattern can restart partway through, and you will find two blanks in a row somewhere in the middle.
• An annexure printed single-sided in the middle of a duplex batch produces its own irregular run.

Look at every even page in the preview before you write anything down. Zoom in on the last two.

3. Write the list. There is no "every second page" option, and no step syntax — you cannot type 2-24 step 2 and have it understood. You type the numbers:

2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24

For anything long, type it in a text editor and paste it in. Twelve numbers is fine by hand; sixty is not, and mistyping one of them means you keep a blank page and lose a real one.

Ranges do work for consecutive runs. If pages 30 through 40 are all blank because someone scanned the back of an annexure, 30-40 is shorter and safer than eleven numbers.

4. Run it, then read the numbers. The result card shows Original 24 pages, Remaining 12 pages, and the final size. Twenty-four minus twelve is twelve. If it says 11 remaining, you deleted one too many and you will know immediately instead of three days later.

5. Flip through the result. The canvas gets an Original / Result toggle with the page counts in the labels. Go through the result once at readable zoom. You are checking one thing: that no page carrying a signature, a stamp or a handwritten correction went with the blanks.

6. Download. The file comes down as pages_removed_ plus your original name. Rename it to something sensible.

Your source scan is untouched throughout — the tool builds a new document from the pages you kept and never modifies the file on your disk. So a mistake means loading the original again, not recovering anything.

What this does to the file size

Usually a lot, and sometimes less than you expect.

A scanned page is an image, and a blank scanned page is a full-size image of nothing — often 200 to 400 KB each, because the scanner encoded all that noise faithfully. Dropping twelve of them typically takes a 9 MB scan down to 5 MB or so. The Output card shows you the real number.

If the target is a portal that caps uploads at 2 MB, removing blanks alone will not get you there and it was never going to. That is a compression job: run the trimmed file through the PDF Compressor, where Low, Medium and High trade sharpness against size, and Screen goes further when a form has a genuinely tight limit. Trim first, compress second — there is no sense spending compression effort on pages you are about to delete.

The other duplex problem: upside-down backs

Sometimes the backs are not blank at all. If the sheets were printed on the long edge but the scanner was set to flip on the short edge, every second page comes out rotated 180 degrees. The content is all there, all readable, and every alternate page is upside down.

Do not delete those. Rotate them. The PDF Rotator has an Apply to switch, and with Specific selected you list exactly the affected page numbers and turn only those by 180 degrees while the rest stay as they are.

Fixing it at the source

Since this will happen again, it is worth thirty seconds at the machine.

If you are scanning single-sided paper, turn duplex off before you start. On most office scanners it is a single toggle on the panel, sometimes labelled 2-Sided or Duplex, and on flatbed software it is a checkbox in the scan dialog. At a cyber cafe, say "single side only" before they begin — most operators leave the last customer's setting in place, which is exactly how you got here.

And if the job produced several files that need joining anyway, do the blank removal on each piece first and then combine them in the PDF Merger. Cleaning up a twenty-four page file is comfortable. Hunting blanks through a merged ninety-page one, where you have lost track of which section each page came from, is not.

Everything here runs in your browser, so a scanned rent agreement or a set of medical records is never uploaded anywhere while you tidy it. The full set of document tools is on the all tools page.