Someone asks you to send "a scan" of a document. You don't own a scanner. You photograph the sheet with your phone, attach the JPEG and hope it'll do. Half the time it comes back: crooked edges, the shadow of your hand, the table included in the frame, a 6 MB file for a single page. The difference between a photo and a scan isn't the camera — it's what happens after the shot.
Why a photo isn't a scan
A flatbed scanner has three structural advantages a camera doesn't: the document is pressed flat, the light is uniform and the distance never changes. The result is a head-on image, evenly lit, with clean edges.
A handheld photo instead introduces three systematic distortions:
- Perspective: if the phone isn't perfectly parallel to the sheet, the rectangle becomes a trapezium.
- Uneven lighting: one side of the page is brighter than the other, and the photographer's shadow is often in the frame.
- Unwanted context: the table, your hand, the edge of a notebook end up in the image.
Then there's file size: a full-resolution photo can take several megabytes per page, and a ten-page document becomes an attachment many mail systems refuse.
What a camera scanner actually does
A scanning tool doesn't just save the shot: it corrects it. Cropping removes the background and brings the document back to its real edges; filters act on brightness and contrast so the text separates cleanly from the paper; export merges multiple pages into a single PDF instead of producing a folder of scattered images.
It's also why a software scanner always uses the rear camera and aims for the highest available resolution: the front camera, designed for selfies, has a smaller sensor and yields insufficient detail on small type.
Which filter to use, and when
| Filter | When to use it | Effect |
|---|---|---|
| None | Colour documents, charts, photos, coloured stamps | Preserves the original colours |
| Greyscale | Ordinary printed text, handouts, notes | Cuts file size while keeping tonal range |
| Black and white | Crisp text on white paper, forms, faxes | Maximum contrast, very light file |
| Enhance | Yellowed paper, poor light, faded photocopies | Raises contrast and brightness without losing colour |
The most common mistake is applying black and white to everything. On a document with a coloured stamp or a blue-ink signature, the conversion threshold can erase precisely the element that makes the document valid. When in doubt, "enhance" is the safer choice.
How to scan, step by step
- Open the free document scanner and allow camera access — the prompt comes from the browser, not the site.
- Place the sheet on a dark, uniform surface: contrast with the background makes cropping far more accurate.
- Frame from above, as parallel to the sheet as possible, in diffuse light. No flash: it creates a central reflection that wipes out the text underneath.
- Take the shot, then crop to the document's real edges and pick the right filter.
- Add the following pages to the same document, reorder them if needed, and export everything as a single PDF.
If you need the document to be searchable — with selectable text rather than just an image — the next step is OCR, which converts the photographed characters into real text.
The mistakes that ruin a scan
- The flash. On glossy or laminated paper it produces a white halo in the middle of the page, and it's unrecoverable.
- A curved page. Photographing an open book without flattening it warps the lines towards the spine: still readable to the eye, but problematic for OCR.
- A light background. White sheet on a white table: edge detection can't find the boundary and cropping has to be done entirely by hand.
- Shooting one page at a time as separate files. Ten JPEGs are unmanageable for whoever receives them; a single PDF is the expected form.
- The shadow of your phone or hand. Move the light source to the side, or change position: it's more effective than any filter applied afterwards.
Frequently asked questions
Do I need to install an app to scan?
No. A browser-based scanner uses the device's camera directly, with the permission you grant at that moment, without installing anything. It works from a phone or from a computer with a webcam, though a smartphone's rear camera almost always gives better results.
Can I put several pages into a single PDF?
Yes, and it's the recommended approach: shoot the pages one after another within the same session, reorder or delete the ones that came out badly, and export a single final file.
Is the scan searchable with Ctrl+F?
Not by itself: the resulting PDF contains images. To make it searchable you need to run optical character recognition, which extracts the text from the image and makes it selectable.
How do I reduce the file size?
By using greyscale or black and white for text-only documents: a form scanned in black and white can weigh a fraction of the same document in colour, with no loss of readability.
In summary
A usable scan doesn't require a scanner: it requires a head-on photo, diffuse light, a crop to the real edges and the filter suited to the document type. The camera document scanner puts these steps together and produces a multi-page PDF directly, instead of a folder of JPEGs. From there, if the document needs to be searchable it goes through OCR, and if you need to merge, rotate or reorder it you can use the PDF page editor; every tool is collected on the PDF and AI tools page.