Split PDF pages in half
Turn scanned book spreads and two-up handouts into single pages. Automatic spine detection, gutter trim and right-to-left reading order.
- Runs in your browser
- Files are never uploaded
- Free, no account
Your scan never leaves this tab — here is how to check
This page has no upload step and no server behind it. Two ways to confirm that yourself, neither of which requires trusting us:
- Turn off Wi-Fi or put the device in aeroplane mode after this page has loaded, then run the tool. It still works.
- Open your browser’s developer tools, switch to the Network tab, clear it, and process a file. No request appears.
The reason this matters: in 2025 the PDF service pdf.net was found exposing roughly 3.5 million user-uploaded documents, and formatter sites JSONFormatter.org and CodeBeautify leaked over 80,000 saved snippets containing live cloud credentials. Those are architectural failures — a file that is never sent cannot be exposed that way.
The problem this fixes
Open a book on a flatbed scanner and every scan captures two printed pages at once. Open a two-up conference handout and every sheet holds two slides. Either way the PDF has half the pages it should, each one twice as wide, and it is unreadable on a phone, an e-reader or a tablet held in portrait.
Splitting each sheet down the middle turns a 180-page spread scan into a 360-page document that reflows properly on any screen and prints one page to a sheet.
Finding the spine automatically
Press Find the spine and the page currently in the preview is rendered at low resolution inside this tab. Every pixel column is scanned and the dark pixels are counted, producing an ink-density profile across the width of the page. The gutter between two printed pages is, almost by definition, the emptiest column near the middle — so the lowest point of that profile inside the central third is where the cut goes.
If the profile has no clear minimum — a full-bleed photograph, a table that runs straight across the spread — the tool says so rather than guessing, and you place the line by hand.
Gutter trim, and why the inner edge looks dirty
A book pressed onto a scanner glass curves away near the spine. That curve shows up as a grey-to-black gradient a few millimetres wide down the inner edge of both halves, and it looks worse after splitting because it is now at the edge of a page rather than in the middle of a spread.
Set Trim at the cut to 3–8 mm and that band is removed from both halves. The trim is symmetrical around the split line, so the two pages stay the same width. Start small: too much trim eats into the inner margin of the text block.
Reading order
For Latin-script books the left half of a spread comes first. For Arabic, Hebrew, Japanese manga and traditional Chinese, the right half comes first. Switch the reading order and the two halves are written into the output in the opposite sequence — the cut itself is unchanged, so you can flip this after the fact without recutting.
The same setting handles the horizontal case: with a top-and-bottom split you choose whether the top or the bottom half leads, which matters for two-up handouts printed in a column.
Why the file does not double in size
Each half is produced by embedding the source page with a clipping box, not by re-rendering it. Both halves therefore point at the same underlying scan image inside the PDF, and only the clipping instructions are new. A 400 MB book scan splits into a document with twice the page count and roughly the same file size, at exactly the original resolution.
Pages that carry a rotation flag are handled correctly too: the cut you draw on screen is mapped back through the rotation, and the halves come out the right way up.
A sensible workflow for a whole book
- Split the spreads here, leaving “only split wide pages” on so the cover survives.
- Check the first and last spread — scanners often shift alignment over a long run.
- If a few pages are upside down, fix them with Rotate PDF.
- Trim the outer edges with Crop PDF if the scan captured the platen.
- Reduce the file for a reader with Compress PDF.
Questions
How does spine detection work?
The page is rendered at low resolution in your browser and the amount of dark ink in every pixel column is counted. The emptiest column near the middle of the page is almost always the gutter between the two printed pages, and the cut is placed there.
My scan is crooked and the split cuts through text — what now?
Spine detection only finds a straight vertical cut. If the scan is skewed, deskew it first in your scanner software, or move the split slider by hand and add a few millimetres of gutter trim so the stray sliver is removed rather than kept.
Does splitting reduce the scan quality?
No. Each half is a clipped view of the original page, not a re-render, so a 600 DPI scan produces two 600 DPI halves. File size barely changes because both halves still reference the same embedded image.
How do I handle Arabic, Hebrew or manga scans?
Set the reading order to “right half first”. The cut is identical; only the order in which the two halves are written into the output changes, so the resulting PDF reads correctly right-to-left.
Why did the cover page come out unsplit?
The “only split wide pages” option is on by default, so any page that is taller than it is wide — typically the cover, the back and single-page inserts — is copied through untouched. Turn it off to split every page in the selection.