Why scans end up with blank pages
Most document scanners with a feeder can scan both sides of every sheet in one pass, called duplex scanning. That is handy for a stack that mixes one-sided and two-sided pages, but every one-sided sheet then adds an empty back to the PDF. A 20-sheet contract with 5 printed backs comes out as 40 pages, 15 of them blank. Separator sheets, cover pages and the last sheet of a fax add more.
Many office scanners and multifunction printers have a setting for this, called blank page skip or blank page removal. Their manuals are open about its limits: a Toshiba manual warns that some pages that are not blank may be detected as blank, and some blank pages may not be detected. This tool does the same job after the scan, but it shows you every decision before the file is saved.
How the detection works
Each page is drawn in your browser at about 600 pixels on its long side. The tool turns every pixel into a gray value from 0 (black) to 255 (white) and counts it as ink when it is darker than 200. Light gray paper, the faint show-through from the back of thin paper and JPEG noise stay above that line. A band of 5% along each edge is left out, because scanners often leave a dark line, a shadow or the edge of the paper there.
The ink share is the number of ink pixels divided by the pixels inside that band. The detection level turns into a limit between 0.01% and 1% of the page, on a log scale. At the middle position the limit is 0.1%: a page counts as blank when less than one pixel in a thousand is ink.
| Slider position | Blank below | Ink pixels allowed on a US Letter page |
|---|---|---|
| Far left | 0.01% | fewer than 23 |
| One quarter | 0.03% | fewer than 72 |
| Middle (default) | 0.1% | fewer than 226 |
| Three quarters | 0.32% | fewer than 714 |
| Far right | 1% | fewer than 2,258 |
Worked example: a US Letter page (8.5 × 11 inches) is drawn at 464 × 600 pixels. Without the 5% band on each side, 418 × 540 = 225,720 pixels are measured. At the middle level, 0.1% of that is 225.7, so the page counts as blank when fewer than 226 of those pixels are ink. An A4 page is drawn at 424 × 600 pixels and measured on 382 × 540 = 206,280 pixels. Every page in the grid shows its own measured share, so you can see how close it is to the limit.
What happens when you save
The tool opens your original PDF, takes the marked pages out and saves a copy. The pages you keep are not drawn again or recompressed. Text stays selectable and searchable, scans keep their full quality, and links and form fields on kept pages keep working. Objects that only the removed pages used, such as the scanned image of an empty back, are dropped from the file, so it gets smaller. Your original file stays unchanged.
Limits
- Very light content. Faint pencil, a light highlighter mark or a pale watermark can stay above the ink threshold. Such a page looks blank to the tool. Check marked pages before you save.
- Dark or colored backgrounds. A scan with a gray or dark scanner lid, or a page on colored paper, is dark everywhere and is never found as blank. Click such pages to remove them by hand.
- Small content. A page with only a page number or a signature line has little ink. Lower the level if those pages must stay.
- Encrypted files. A PDF with a password or editing restrictions cannot be changed here. Save a copy without protection first.
- Memory. Pages are measured one at a time, so long documents work, but a PDF of several hundred megabytes can be too much for a phone or tablet.
To remove pages by number instead, use Delete PDF Pages. To put the remaining pages in order, use Reorder PDF Pages. A clean scan is easier to search after OCR PDF adds a text layer, and Compress PDF makes it small enough to email.
How to remove blank pages from a PDF
- 1Drop your PDF on the box or choose it from your device. Every page is drawn and measured, and blank pages get a red frame.
- 2Move the detection level until the right pages are marked. Click any page to keep or remove it.
- 3Check the counts, then press Save PDF without blank pages. The new PDF downloads with "-no-blank-pages" in its name.
Sources
- Toshiba manual, Eliminating blank pages from transmission (blank page skip): blank page skip for two-sided scanning, and the note that some pages that are not blank may be detected as blank and some blank pages may not be detected
- Konica Minolta bizhub manual, Blank Page Removal: blank pages in the feeder are skipped, and the detection level can be adjusted when blank pages are not detected correctly
Related guides
- How to Fix a Scanned PDF: The Right Order of Steps
- Email Attachment Size Limits, and How to Get Under Them
- How Long to Keep Tax Records: The IRS Periods, With Dates
Frequently asked questions
How does the tool decide that a page is blank?
It draws each page small, counts the pixels that are clearly darker than paper, and leaves out a thin band along the edges where scanners leave shadows. If the dark pixels cover less of the page than the detection level allows, the page is marked as blank. You see the ink share of every page and can change any mark by hand.
Will it remove pages that are almost empty?
It can. A page with only a page number, a short note or a faint pencil mark has very little ink, and at a high detection level it counts as blank. That is why every marked page is shown before you save. Lower the level or click the page to keep it.
Does the text in my PDF stay selectable?
Yes. The pages you keep are not redrawn. The tool takes them out of the original file as they are, so text, links on kept pages and the image quality of scans stay exactly as before. Only the removed pages are cut out.
Is my PDF uploaded?
No. Your browser reads the PDF, measures the pages and saves the new file on your device. Nothing is sent to a server.
Why is a dark or colored page not found?
The tool looks for paper that is close to white. A scan with a gray or dark background, or a page printed on colored paper, has dark pixels everywhere, so even an empty one looks full of ink. Mark such pages by hand with a click.
Can I get the removed pages back?
Not from the new file, but your original stays as it was. The tool saves a copy with "-no-blank-pages" in the name and never changes the file on your device.