Scanning a book almost always produces spreads. The book lies open, one capture takes in both facing pages, and you end up with a PDF of wide sheets that is fine for archiving and miserable to read — particularly on a tablet, where every sheet has to be zoomed and panned twice. The fix is to cut every page down the middle so the file becomes ordinary sequential pages.
How to split a scanned book into single pages
- Open ihatepdf.cv/split-pdf-in-half — nothing to install, and the scan is never uploaded.
- Drop in your scanned PDF. The first page is rendered as a preview.
- Choose "Each page in two" and a vertical cut.
- Drag the cut position until the guide line sits on the spine — scans are rarely perfectly centred.
- Split and download. Every wide sheet becomes two pages, left half first, so the book reads in order.
Getting the cut in the right place
Flatbed and phone scans are rarely square. The spine drifts off centre, the book curves away from the glass, and the two halves end up slightly different widths. A fixed 50 per cent cut then clips one page and leaves a strip of its neighbour on the other.
An adjustable cut position solves this: set it against the preview so the line lands in the gutter rather than on text. If the drift changes through the document — common when a book was scanned in one sitting and the binding gradually relaxed — split in batches, adjusting for each run, and merge the results afterwards. For pages that also need their edges tidied, run the output through Crop PDF.
Why the order comes out right
The one thing a splitter must get right is sequence. Cutting each sheet produces two pages, and they have to be emitted left half first, then right half, before moving to the next sheet. Get that wrong and a 200-sheet scan becomes 400 pages in an order nobody can read. It is worth spot-checking the first four pages of any output before you commit to it.
Cropping is not rasterising
It matters how the halving is done. Each half should be produced by changing the page box — the window through which the page is viewed — rather than by rendering the page to an image and cutting the picture up.
Every consequence of that is good. Text stays selectable and searchable if the scan has been through OCR. Vector artwork stays sharp at any zoom. Quality is untouched because nothing is re-encoded. And the file does not balloon, which a rasterising tool will reliably do.
There is one consequence to be aware of: the content outside the visible half still exists in the file. For a book scan that is irrelevant. For anything where the hidden half must genuinely be gone, cropping is the wrong tool and Redact PDF is the right one.
Splitting the document versus splitting the pages
"Split in half" means two different things and it is worth being clear which you want. Splitting the document in half divides it at a page boundary — a 40-page PDF becomes pages 1–20 and 21–40 as two files. Splitting the pages in half divides every sheet itself, which is what book spreads need. They solve unrelated problems and both live on the same page here.
Horizontal cuts have their uses too
Vertical is what book scans need, but a horizontal cut across the middle suits pages holding two stacked items: certificates, tickets, vouchers, address labels, or forms printed two to a portrait sheet. The mechanism is identical — only the axis changes.
What to do after splitting
Two follow-ups are usually worth it. Run the result through OCR if you have not already, so the book becomes searchable — this is far more useful on single pages than on spreads, because search results land on the actual page. And consider compressing, since scanned books are almost entirely image data and typically shed half their size with no visible change.
Why not to upload a scanned book
Scanned books are frequently under copyright, and personal scans — journals, letters, family documents — are frequently sensitive. Uploading them to an online tool means handing a complete copy to a third party. Splitting in the browser avoids the question entirely: the file is opened, cut and saved on your own machine, and once the page has loaded you can disconnect from the network and it still works.
Frequently asked questions
How do I separate a scanned book into single pages?
Choose "Each page in two" with a vertical cut. Every wide sheet becomes two pages, left half first, so the output reads in the correct order from beginning to end.
The spine is not in the middle of my scan. Can I still split it?
Yes — that is what the cut position slider is for. Drag it until the guide line sits on the spine in the preview.
Is any content lost at the cut?
No. Each half is a crop window over the original page, so everything is preserved and text stays selectable.
Does cropping actually remove the other half from the file?
No. Cropping hides content rather than deleting it, so it is not a redaction. Use Redact PDF if content must genuinely be removed.
Will the split pages still be searchable?
Yes, if the scan has a text layer from OCR. Cropping does not disturb it. If it has no text layer, run OCR after splitting.
Can I split at a point other than the exact middle?
Yes, anywhere between 10 and 90 per cent of the page width or height.
What about a horizontal split?
Also supported — useful for certificates, tickets and forms printed two to a sheet.