fixmint
EN

How to scan a document with your phone and get a clean PDF

By · · 4 min read

To scan a document with your phone, put the page flat on a plain, contrasting surface in even daylight, photograph it from directly above so the page fills the frame, then straighten the image, convert it to black and white and save it as a PDF. Most of the quality comes from light and flatness, not from the camera's megapixel count.

Scan documentPhone photo of a page looks badDo it free

Why phone scans look poor

Most bad phone scans have one of four faults. A shadow falls across one side of the page, usually from the phone or your hand. The page is photographed at an angle, so the lines of text run into each other and the page becomes a trapezoid. The paper curls, especially near a book's spine or a stapled corner, and the curl shows as shading. Or the photo is blurred by hand movement in dim light.

Fix the light and the angle first. Filters come last.

Light, surface and angle

Use daylight from a window, and keep the light even. Turn off the flash: it makes glossy paper glare and leaves a bright patch over the text.

Put the page on a surface that contrasts with it. A white page on a dark table is ideal, because the edges of the paper are easy to find. Keep the page flat and smooth, and hold the phone above it with the lens roughly over the centre of the page. Keep your fingers and the shadow of your hand out of the frame.

The scan tool finds the four corners of the page for you. If it puts a corner in the wrong place, drag it. Once the corners are right, the tool straightens the page into a rectangle, which is what corrects the trapezoid.

Resolution: what 300 dpi means

A scan's resolution is the number of pixels per inch on the paper. The U.S. federal digitisation guidelines (FADGI) use 300 pixels per inch as the baseline for printed text, and OCR software works best at that level or higher. At 300 dpi, a US Letter page of 8.5 × 11 in is 2,550 × 3,300 pixels. An A4 page of 210 × 297 mm is 2,480 × 3,508 pixels.

A phone photo of a full page usually has more pixels than that. The number that matters is how many pixels cover the text. If the page is only a small part of a wide photo, the text may have fewer pixels than you think, so fill the frame with the page.

Black and white, greyscale or colour

The three filters trade detail against file size:

  • Black and white turns every pixel into black or white. It gives sharp text and the smallest file. It can lose faint pencil, light stamps and fine shading.
  • Greyscale keeps shades of grey. It suits pages with photos, logos, faint print or a pen signature.
  • Colour is needed when colour carries meaning, such as a coloured stamp, a highlighted clause or a chart.

The file size difference is easy to work out. A Letter page at 2,550 × 3,300 pixels has 8,415,000 pixels. Stored uncompressed at one bit per pixel, which is black or white only, that is 8,415,000 ÷ 8 = 1,051,875 bytes, about 1.05 MB. Stored at eight bits per pixel, which is greyscale, it is about 8.4 MB before compression. Compression shrinks both, but black and white starts far smaller.

Pick black and white for a plain letter. Then check the faint parts: a light signature, a small footnote, or a pencil note. If one of them has gone, scan again with greyscale.

Several pages, in order

Scan the pages in reading order and add them to one PDF. Check the page order when you open the finished file, and check the last page too, since it is easy to lose one. If a page is in the wrong place, use the reorder tool rather than scanning the whole document again. If you already have photos of the pages and want to skip the scan step, images to PDF combines them in order.

Email and upload limits

Gmail's help page says personal accounts have a 25 MB limit for attachments, and the limit covers the total of all attachments in one message. If the total is over the limit, Gmail adds the files to Google Drive and puts a link in the email instead. Other providers and upload forms set their own limits, so check the one you have.

If the PDF is too large, try a black and white filter before anything else, because it changes the file far more than a small resize would. Then run the compress tool for PDFs. After compressing, open the PDF at full size and check that small text is still sharp.

A scanned PDF is a picture of the page, so its text cannot be selected or searched. If you need the text itself, run the page through the image to text tool first.

Before you send

  1. Check that the corners were found and the page is square.
  2. Check the faintest text, the signature and any stamp.
  3. Check the page count and order.
  4. Name the file so it says what it is and when it was made, and keep the original scans.

Quick answers

Why does my scan come out grey? The paper colour and any shadow both turn grey when photographed. A black and white filter removes the tint.

Is a phone scan as good as a flatbed scanner? For text on a flat page in good light, a careful phone scan is readable. A flatbed scanner gives more even results for photos and fine detail.

Why is the PDF so large? Usually because the pages are in colour or greyscale. Switch to black and white for text-only pages, then compress if it is still too big.

Comments

    No comments yet. Be the first.

    Your rating (optional)

    Your email is never shown or kept.

    Cover of Digital Workflow ToolkitDigital Workflow Toolkit

    Recipes, templates and patterns for chaining small file jobs (convert, rename, compress, merge) into workflows that run the same way every time.

    $19.00Get it on Gumroad