Document Scanner

Scan documents using your camera with auto-crop, perspective correction, and PDF output.

Tested tool guide Tested browser tools Checked August 16, 2026

What Document Scanner does and how it behaves

Document Scanner captures a paper page through your device camera, removes the surrounding scene with auto-crop, corrects the apparent shape of a page photographed at an angle, and produces a PDF. Camera access requires browser permission, and the document is handled entirely in the browser rather than uploaded. The important limitation is that geometric correction cannot repair missing, blurred, shadowed, or glare-obscured content. A straight-looking scan is not necessarily a readable one.

How the result is produced

Capture and crop

After camera permission is granted, place the complete document within the camera view. Auto-crop identifies the apparent page boundary and excludes the visible desk or other background around it. The boundary is easiest to distinguish when all corners are visible and the paper contrasts with its surroundings. Review the captured area carefully because weak contrast or overlapping objects can produce an incorrect crop.

Correction and PDF output

Perspective correction reshapes a page photographed at an angle so that its edges appear rectangular instead of trapezoidal. The corrected scan is then available as a PDF document. This operation changes the page geometry, not the captured detail. It cannot reveal text outside the frame, remove motion blur, or reconstruct information hidden by glare, fingers, folds, or deep shadows.

Good uses

Turn a photographed receipt into a tightly cropped PDF for an expense record.
Capture a signed paper page when a flatbed scanner is not available.
Digitize a letter or printed form while removing the desk visible around its edges.

Limits and checks

Inspect every edge before relying on the scan. Auto-crop may cut into a page or retain background when the boundary is faint.
Zoom in on small text. Perspective correction can straighten the page but does not make an out-of-focus or low-resolution capture sharp.
Do not assume PDF output means searchable text. A PDF can contain only a page image, so text selection and search require separate character recognition.

Common questions

Does the scanner recognize or extract the text on the page?

Not from the stated feature set. Auto-crop and perspective correction operate on the photographed page; neither feature implies character recognition. PDF output can contain page images without selectable or searchable text. Treat the result as a visual scan unless the generated document itself provides usable text selection.

Why is part of the page missing or still distorted?

Place the entire sheet inside the camera view, keep the camera as parallel to the page as practical, and use even lighting with visible contrast around every edge. Auto-crop can misidentify a boundary when the page blends into its background. No, perspective correction cannot restore corners or words that were outside the captured frame.

References and verification

The behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.