I have a 400-page scan — is rescanning or splitting faster?

Splitting is faster by an incomparable margin — splitting a 400-page file takes minutes, while rescanning the same book takes hours. Rescanning only makes sense if the existing scan's quality is unusable (very low resolution, heavy skew, too dark to read). If the two-page spread structure is the only problem, splitting with software is the right call.

Can't I just crop instead of splitting and read it that way?

Cropping picks a single area and applies it to every page, so you'll see either only the left pages or only the right pages — you lose half the book. Cropping the book twice (once left, once right) to produce two files and then interleaving them is theoretically possible, but that means doing in three operations what splitting does in one.

Do phone book-scanning apps solve this problem?

Many of them do — they come with page detection, automatic splitting and skew correction. If you're about to make a new scan, those apps are a good starting point. But if you already have a two-page-spread PDF, those features won't help you; to fix that file you still need to split it with software.

Will the split file be lower quality than the original?

No, it won't. Splitting doesn't re-encode the image data; it divides the page's visible area in two. There is a psychological effect though: because each page now appears twice as large on screen, you see the same pixels from closer up, and on low-resolution scans that reads as pixelation. That isn't loss caused by splitting — it's the scan's already existing limit becoming visible.

Split the Spread, Rescan, or Just Crop? Which Makes More Sense

You have a scanned book PDF with two pages on every sheet. It doesn't read on a tablet, it prints at half size, and it doesn't work at all on an e-reader. You have three options in front of you: split it in half with software, rescan the book, or try to get by with cropping. This article compares all three realistically.

The three approaches in brief

Software splitting: takes your existing PDF, cuts every sheet down the middle, and gives you a new file with double the page count. It doesn't touch the image data.

Rescanning: you take the book back to the scanner and this time scan each page individually, or use an app that splits automatically.

Getting by with cropping: without splitting the file, you try to make it readable by narrowing the visible area.

Comparison table

| Criterion | Software splitting | Rescanning | Cropping | |---|---|---|---| | Time (400 pages) | Minutes | 3-6 hours | Minutes | | Requires physical access to the book | No | Yes | No | | Quality loss | None | None (new scan) | None | | Fixes skewed pages | No | Yes | No | | Removes the gutter shadow | No | Partly | Partly | | Center offset problem | Yes (crop first) | No | — | | Gives you the whole book | Yes | Yes | No | | Subsequent OCR quality | Good | Best | Poor |

Scenario 1: I no longer have access to the book

You returned the library book, got the scan from a friend, or downloaded it from an internet archive.

Winner: software splitting, no contest. Rescanning isn't an option. Splitting neither improves nor degrades the existing image quality — it just fixes the structure. In this scenario there's nothing to decide.

The only pre-check you need: whether the cut line will fall in the right place. Look at a few pages and see whether the book's spine sits exactly in the middle of the sheet. If there's a noticeable offset, first trim the excess edge with a crop tool, then split.

Scenario 2: I have the book, but the scan quality is bad

The scan is 150 DPI, the text is blurry, some pages are dark.

Winner: rescanning. Splitting doesn't improve a bad scan — it actually makes the problem more visible. Because a split page appears twice as large on screen, the pixelation feels more pronounced.

When rescanning, watch these settings:

Resolution: 200-300 DPI is enough for reading only. If you're going to OCR, 300 DPI is the floor. On small type or footnote-heavy text, going up to 400 DPI makes a difference.
Color mode: for plain text, grayscale is both sufficient and gives a much smaller file. If there are color figures or photos, scan in color. Black-and-white (1-bit) mode shrinks the file a lot but destroys grayscale figures.
One page per scan: scan each page separately if you can; the problem is solved at the source. If the book is tightly bound and single-page scanning is hard, scanning spreads and splitting with software is still acceptable.

The time reality: scanning a 400-page book by hand, at 20-30 seconds per page, takes 3-6 hours. That's a serious investment and only makes sense if you'll use the file for a long time or you're archiving it.

Scenario 3: I only need to look at a few pages

You need to read three pages from a book and you have no long-term relationship with the file.

Winner: none of them — just zoom in. We put this scenario on the list because people tend to do unnecessary work. Processing a file for three pages is a waste of time; zoom in your PDF viewer and read.

If you do need to prepare those three pages properly for someone, first pull out the relevant sheets with an extraction tool, then split only those. That's far faster than processing 400 pages.

Scenario 4: I'm moving it to an e-reader

You'll be reading on a Kindle, Kobo or similar device.

Winner: software splitting plus extra steps. E-readers have small screens and two-page-spread PDFs are practically unreadable there. Splitting is mandatory.

But splitting alone may not be enough. To improve the e-reader experience:

Split — so each page stands alone.
Crop — throw away the margins; every millimeter counts on an e-reader screen. The gutter shadow also gets largely cleaned up in this step.
Compress — e-readers have limited processing power; page turns get slow on large files.

After those three steps a split scan becomes genuinely usable on an e-reader.

Scenario 5: I want to search the text (OCR)

You want to be able to search the document and copy quotations.

Winner: split first, then OCR. And the order is critical.

On a two-page sheet the OCR engine sees two columns and has to decide how to order them. Sometimes it guesses right; sometimes it reads lines horizontally left to right and interleaves the text of two pages. The result is an unreadable text layer and search that doesn't work.

On split single pages there is no such ambiguity.

If the scan quality is low and the OCR result comes out bad, rescanning may genuinely be needed here. What OCR notices most is resolution and contrast; the recognition accuracy gap between a 150 DPI scan and a 300 DPI scan is large.

Scenario 6: I'm about to make a new scan — how should I do it?

You haven't scanned yet and you want to prevent the problem from the start.

Best: scan each page separately. The problem never appears.

Practical alternative: use a phone app that splits pages automatically. Many of these apps detect page edges, correct skew and separate two pages automatically. On thick hardcover books they can beat even a flatbed scanner, because they flatten from angled photos without forcing the book open.

Acceptable: scan spreads and then split with software. If you take care to center the book, the split comes out clean.

Decision summary

Let's simplify:

If you don't have the book → split, no other option.
If you have the book and the scan quality is good → split; rescanning is a waste of time.
If you have the book and the scan quality is bad → rescan, especially if you're planning OCR or long-term archiving.
If you only need a few pages → extract first, then split just those.
If you haven't scanned yet → scan pages separately or use an app that splits automatically.

Getting by with cropping wins in almost no scenario, because a single crop area is applied to every page and makes half the book invisible. Cropping is not an alternative to splitting but a complementary step done before it (to fix the center) or after it (to clean up the gutter shadow).