Shot2Text

How to Extract Text From an Image (Every Method, Ranked)

Paste, file, paper, PDF — the right method depends on what kind of image you have. Here they are in order of speed, with the cases where each one breaks.

There is a picture of some text, and you want the text itself — selectable, searchable, and yours to edit. This is the procedure, in order, including the cases where each method fails.

The fastest method: paste it into a browser tab

If the image is on your screen or already on your clipboard, this takes about four seconds.

  1. Copy the image — screenshot it, or right-click and Copy image.
  2. Open the screenshot to text tool.
  3. Press Ctrl + V (or ⌘ + V on a Mac).

The text appears as plain text you can select and copy. Nothing is uploaded, so there is no waiting for a file to travel and nothing stored on a server afterwards.

This is the right method for: error messages, code in a tutorial, a paragraph you want to quote, numbers from a dashboard. Anything that is already pixels on your screen.

If the image is a file on disk

Drop the file onto the tool instead of pasting. JPG, PNG, WebP and BMP all work. Same result, same speed.

One thing worth knowing: if you have a choice of format, PNG gives the best accuracy. It is lossless, so the letters arrive with clean edges. A JPG has compression halos around every letter — invisible to you, but they look like part of the letter to a reader. If you exported a screenshot as JPG, re-export as PNG and you may find the errors disappear entirely.

If the image is on paper

Photograph it rather than scanning it. Phone cameras are far better than most people expect, and the photo to text tool corrects for the washed-out grey that indoor light produces.

The rules that matter, in order:

  • Straight down, square to the page. A few degrees of tilt starts losing whole rows.
  • Fill the frame with the text. Not the desk, not the margins — the words you actually need.
  • Light from behind or beside you. Your own shadow across the page is the single most common cause of a failed conversion.
  • No flash. It creates a hotspot in the middle and a dark edge, which is exactly what the reader handles worst.

If the image is inside a PDF

Then you have one of two documents, and the right method depends on which.

Open the PDF and try to select a word with your cursor. If the text highlights, it is a digital PDF — the letters are stored in the file. Do not run it through OCR; just copy. Text extracted from the file itself is exact, with zero errors. Running OCR on a digital PDF is strictly worse.

If nothing highlights, it is a scanned PDF — a set of photographs bound into a file. There are no letters in it to select. Use the scanned PDF to text tool, which renders each page and reads the words back out.

When the result comes out wrong

Almost every failure traces back to the image, not the reader. In rough order of frequency:

The text is too small. Below about 12 pixels per character there is nothing to recognise. Crop to the text, or zoom before screenshotting.

The contrast is too low. Pale grey on white, or pale grey on near-black. Increase contrast, or capture the light-mode version of the page.

The image has been recompressed. Anything sent through a chat app has been shrunk and requantised. Use the original file from your camera roll, not the version that arrived in a message.

Text sits on a busy background. The reader cannot separate letters from texture at similar brightness. Crop tightly.

The layout has columns. The reader scans line by line across the full width, so two columns merge into one stream. Crop one column at a time.

What you get, and what you don't

You get the characters, in reading order, with line breaks. You do not get bold, italics, headings, fonts, colours, tables or images. Formatting is not visible to a text reader — it sees shapes, not styles.

If you need a document rather than text, the image to Word converter builds a real .docx with one paragraph per line, which is a much better starting point for editing than a wall of pasted text.

A note on privacy

Most free extraction services work by uploading your image to a server. That means a copy of your screenshot — a receipt, a chat, a document — sits on someone else's disk, usually for an unspecified length of time.

Every tool on this site runs the recognition inside your browser tab. The image is never transmitted. You can verify it rather than take our word for it: open your browser's developer tools, switch to the Network panel, and run an image through. You will see the English text data download once. You will not see a request carrying your image.