How to Extract Text From an Image

First, a straight answer about this site

We convert between image formats. We do not do OCR, and we are not going to pretend otherwise — a good OCR engine is a large machine-learning model, not a canvas operation, and shipping one into a browser tab would mean a download of tens of megabytes for a job your device already does well.

What follows is where to find the tools that do it properly, all of them free, most of them already installed.

On an iPhone or iPad

Live Text works in Photos, Camera, Safari, Quick Look and screenshots, in a long list of languages. It also recognises phone numbers, addresses and dates and offers to act on them. For a single page of text it is the fastest option on any platform.

1. Open the photo in Photos, or point the Camera at the text.
2. Press and hold on the words. A selection appears with handles.
3. Drag to select what you need, then Copy — or tap the Live Text icon in the corner to select everything at once.

On Android

Google Lens also works live through the camera and from the Google app, and the accuracy on printed text is excellent. Some manufacturers add their own equivalent in the gallery app, which is usually the same engine.

1. Open Google Photos and the picture, then tap Lens.
2. Tap Text, then Select all — or drag to choose part of it.
3. Copy, or send straight to your computer with Copy to computer if you are signed in on both.

On a Mac

This works on screenshots, photos and PDFs of scans. For a scanned PDF that needs to become searchable rather than just copyable, Preview can export to PDF with text recognition, and Adobe Acrobat does a more thorough job.

1. Open the image in Preview, or just select it in Finder and press Space for Quick Look.
2. Drag across the text as if it were a document. Live Text makes it selectable in place.
3. Copy and paste anywhere.

On Windows

• Snipping Tool: take a capture, then use the text actions button to copy all the text or just the email addresses and phone numbers.
• PowerToys Text Extract: a free Microsoft utility that grabs text from anywhere on screen with a keyboard shortcut. The best option if you do this often.
• OneNote: paste an image, right-click, Copy Text from Picture. Long-standing and reliable.
• Microsoft Lens on a phone: photograph a document and export it as editable Word text.

For whole documents

The built-in tools are aimed at a screenful at a time. For a multi-page document you want something that keeps the structure.

• Google Docs: upload the image or PDF to Drive, right-click, Open with Google Docs. It OCRs the file and gives you an editable document with the image above the text. Free, handles many pages, and surprisingly good.
• Microsoft OneDrive and Word: uploading a PDF and opening it in Word runs recognition and rebuilds the layout.
• A dedicated scanner app: most phone scanner apps run OCR as they save, which is the least work if you are digitising a filing cabinet.

Getting better results

That fifth row is where our black and white converter genuinely helps: running a photographed page through a threshold conversion before OCR often lifts accuracy noticeably, because it removes paper texture and show-through from the other side of the page.

A word about privacy

The on-device options — Live Text on Apple platforms, PowerToys and Snipping Tool on Windows — process the image locally. Google Lens and Google Docs send it to Google's servers. For a photographed payslip, medical letter or ID document, that distinction is worth caring about, and the on-device route is the one to use.