Clean crooked and washed-out PDF scans
Auto-contrast, deskew and optional despeckling on every scanned page.
Free: 3 jobs per UTC day, 25 MB per file. PDF exports include a visible Made with Bindery watermark. Pro exports clean PDFs.
Prefer the full studio? Open Clean Scanned PDF in the app →
Bindery analyzes each rendered scan locally, lifts weak contrast, estimates small page rotations and removes isolated dust pixels when requested. The output rebuilds the document from cleaned page images, which makes inconsistent scans easier to read and prepare for OCR.
Cleanup rasterizes the pages and removes an existing text layer. Run OCR afterwards if the cleaned result must be searchable, and keep the original for archival or evidentiary use. Bindery runs entirely in your browser — your file is never uploaded, so there is nothing to delete from a server afterwards.
How it works
- Drop the scan and open Scan cleanup
- Choose auto-contrast, deskew and optional despeckling
- Clean the pages, inspect the output and run OCR if searchable text is needed
What scan cleanup fixes
Scanned pages have three common faults: they are slightly rotated because the sheet was not square on the glass, they are grey and low-contrast because the scanner exposed for the paper, and they carry dust — isolated dark pixels that look like noise and confuse OCR. Bindery analyses each rendered page on your device, estimates and corrects small rotations, stretches the contrast so paper reads white and ink reads black, and removes isolated specks when despeckling is on. The result is a document that reads better on screen, prints cleaner, and recognises far more accurately in OCR.
What cleanup changes about the file
Cleanup rebuilds pages from the cleaned images, so it rasterises the document and drops any existing text layer. That is the right order of operations for a scan — clean first, then OCR — but it makes cleanup wrong for born-digital PDFs with real text, which need no cleaning at all. Deskew targets errors up to about three degrees; a page scanned sideways needs Rotate first. Keep the original scan for records, evidence or archives: cleanup is for a readable working copy, and pixels do change.
Cleaning in the browser versus uploading
| Clean scanned PDF | Bindery | Upload-based tools |
|---|---|---|
| Where the file is processed | In your browser tab, on your device | Uploaded to the service’s servers |
| Steps | Auto-contrast, deskew, optional despeckle | Varies |
| Records and IDs | Never leave the device | Uploaded for processing |
| What exists afterwards | The original and the new file, both on your device | A server copy until the retention window ends |
| Works offline | Yes, once the app has loaded | No |
A good scan pipeline
- Capture at 300 DPI, or with Scan from phone, which straightens pages as you shoot.
- Rotate any sideways pages and delete blank ones.
- Clean: contrast, deskew, despeckle.
- OCR in the document’s language.
- Compress — the Balanced target keeps scanned text legible while shrinking the file substantially.
Questions people ask
How much rotation can deskew correct?
The current detector targets small scan errors up to about three degrees. Larger rotations should be fixed with page rotation before cleanup.
Will cleanup remove the searchable text layer?
Yes. Cleaned pages are rebuilt as images, so run the OCR tool afterwards to add a new searchable text layer.
Should I keep the original scan?
Yes. Cleanup changes pixels and is intended for a readable working copy. Keep an untouched original for records, evidence or archival use.
Will cleanup fix a blurry scan?
No. Contrast and deskew help legibility, but blur is missing information. Rescan at higher resolution or with better focus.
Does cleanup work on colour scans?
Yes. Contrast is lifted while colour is kept. For pure text documents a greyscale or black-and-white rescan compresses far better afterwards.
Should I compress before or after cleaning?
After. Clean the best copy you have, run OCR, then compress the searchable result once.