Scan a document from a photo: shoot the paper with your phone, drag four corners onto the page’s edges, and the perspective is corrected so the result looks scanned rather than photographed. A cleanup mode then evens out the paper and the text: Boost for a crisp grey scan, B & W for high-contrast text that copes with a shadow across the page, Grayscale, or Original to keep the colour. One page downloads as a JPG; several become one PDF or a ZIP of JPGs. Photos are processed on your device, which suits IDs, receipts and signed forms.
How it works
- Drop one or more photos of paper, up to 50 pages. Shoot from roughly above, in even light, with the whole page in the frame.
- Drag each corner handle onto a corner of the page; they start near the photo’s edges. Pick an enhance mode, press Apply, and hold Compare to check against the photo.
- Download: one page saves as a JPG at the quality you set (92% to start); more pages export as one PDF, or as a ZIP of JPGs.
Questions
- How do I scan a document without a scanner?
- Photograph it with any phone or camera, then drop the photo here. Dragging the four corners onto the page tells the tool where the paper is; it maps that shape to a flat rectangle, up to 4,000 pixels on the long edge, and cleans up the tone. You get a straight, evenly toned page instead of a tilted photo of one.
- Which enhance mode should I use?
- Boost, the default, stretches the contrast in grey tones and keeps pencil marks and printed photos. B & W turns each pixel black or white by comparing it with its own neighbourhood, so a shadow does not swallow the text; it suits typed pages. Grayscale only removes colour. Original keeps the colour: pick it for stamps, blue-ink signatures and coloured forms.
- Is the scanned PDF searchable?
- No. The PDF holds one image per page and no text layer. To copy or search the words, run the scan through the image to text tool, which reads it with on-device OCR, or put the finished PDF through PDF to text.
- Does my file get uploaded?
- No. The models and all processing run inside your browser on your own device. You can verify this yourself: open your browser’s DevTools, watch the Network tab, and process a file: there are no uploads. After the first visit most tools also work fully offline.
- How is this different from just cropping?
- A crop keeps the photo’s perspective, so a page shot at an angle stays trapezoid. This tool remaps the four corners to a true rectangle (a perspective transform), which is what makes the result look scanned rather than photographed.
- Does it find the page automatically?
- No. The corners start just inside the photo’s edges and you drag them onto the page, because your eyes beat an edge detector on busy backgrounds, and the export is only ever what you confirmed.