No server upload
No server upload — fully browser-based
UtilityKit
500+ fast, free tools. Most run in your browser only; Image & PDF tools upload files to the backend when you run them.
Document Scanner
Perspective-correct a photo of a document by dragging corner handles. Flatten, enhance, and download as PNG.
About Document Scanner
Document Scanner helps you turn a crooked photo of a document, receipt, or whiteboard into a clean, flat scan. Upload your photo and the tool displays four draggable corner handles. Drag them to align with the document's actual corners, choose an enhancement (color, grayscale, or B&W scan look), and download the corrected PNG. Everything runs in your browser using canvas — no server upload.
Why use Document Scanner
Manual corner control for
Manual corner control for precision correction
B&W scan mode for
B&W scan mode for clean printed-document look
Free with no watermarks
Works on receipts
Works on receipts, books, whiteboards, and forms
No server upload
No server upload — fully browser-based using HTML Canvas, ideal for confidential receipts and contracts.
How to use Document Scanner
- Upload a photo of your document
- Drag the four yellow corner handles to the document corners
- Fine-tune the corner positions for best alignment
- Choose enhancement: None, Grayscale, or B&W Scan
- Click 'Apply & Download PNG'
- Upload a photo of your document — JPG, PNG, WebP, or HEIC (where the browser supports it).
- Drag the four yellow corner handles to the actual corners of the document in the photo.
When to use Document Scanner
- Scanning receipts or invoices photographed at an angle
- Fixing skewed photos of printed documents
- Creating clean scans from book or whiteboard photos
- Correcting perspective on text documents before OCR
- Digitizing paper forms photographed with a phone
- Scanning receipts or invoices photographed at an angle for expense reports.
Examples
Receipt scan
Input: JPG photo of a long supermarket receipt photographed at an angle on a wooden table
Output: Flat, B&W-thresholded PNG of the receipt rotated upright and ready for expense reporting or OCR.
Whiteboard capture
Input: PNG photo of a meeting whiteboard with marker text, taken at a slight side angle
Output: Perspective-corrected grayscale PNG with the whiteboard rectangular and readable, ready to share with the team.
Book page
Input: JPG of a paperback book page photographed leaning back in a chair
Output: Deskewed full-colour PNG of just the page contents, ready for OCR into searchable text.
Tips
- Photograph documents on a contrasting background — a dark table for white paper makes corner alignment much easier.
- Hold the camera as parallel to the document as possible; even a perfect homography cannot recover detail lost to extreme angles.
- Use B&W scan mode when you plan to OCR the result — high-contrast black-on-white is what Tesseract is trained on.
- If the document curls (e.g. a magazine page), flatten it under glass before photographing — perspective correction can't fix paper curvature.
- For receipts, drag the corner handles slightly outside the paper edge to leave a small white border in the output.
- Check the alignment guides before clicking Apply — a misplaced corner can produce a sheared or stretched result that's hard to undo without re-uploading.
Frequently Asked Questions
How does perspective correction work?▾
You place four corner handles on the corners of your document. The tool computes a perspective transform (homography) that maps those corners to a flat rectangle.
Is my photo uploaded anywhere?▾
No. All processing happens in your browser using the HTML Canvas API. Your image stays on your device.
What is B&W scan mode?▾
B&W Scan converts the image to grayscale then applies a threshold — pixels above 50% brightness become white, below become black. This gives a clean, printed-document appearance.
What file formats are supported for upload?▾
Any image format your browser supports: JPG, PNG, WEBP, HEIC (on supported devices). The output is always PNG.
Can I use it on mobile?▾
Yes. Drag the corner handles using touch. The interface is touch-friendly.
What if my document has curved edges?▾
This tool corrects perspective (flat surface photographed at an angle) but not paper curl or bowing. For curved documents, straighten the paper before photographing.
Does it auto-detect the document edges?▾
The corners start at the edges of the image. For best results, photograph your document with some background visible so you can drag the handles to the true document corners.
Explore the category
Glossary
- Perspective transform
- A 3×3 matrix transformation that maps a quadrilateral region in one image plane to a rectangle in another, correcting for camera tilt and viewing angle.
- Homography
- The mathematical name for a perspective transform between two image planes; computed from four point correspondences (your four corner handles).
- Edge detection
- An algorithmic family (Sobel, Canny, etc.) for finding sharp brightness transitions in images; used in some scanners for automatic corner detection — this tool uses manual corner placement instead for precision.
- Bilinear sampling
- A pixel-resampling technique that blends the four nearest source pixels when warping; used to produce smooth output from the perspective transform.
- Threshold (binarisation)
- A processing step that converts a grayscale image to pure black and white by mapping pixels above a brightness cutoff to white and below to black; used in B&W scan mode.
- Canvas2D
- The browser API used to draw and resample pixels in JavaScript; this tool uses it to perform the perspective warp without leaving the page.
- Aspect ratio
- The ratio of width to height of the output rectangle; chosen by the tool based on the corner handle positions to preserve the document's original proportions.