How to Extract Text from an Image (Free, On Any Device)
The fastest way to extract text from an image depends on where the image is and how private it is. There are six good methods, from built-in phone features to browser OCR that never uploads your file, and one important fork in the road: plain text tools scramble tables and struggle with handwriting, so those two cases need a different approach, covered at the end.
Method 1: A free browser tool (works everywhere, nothing uploaded)
The most universal option is a browser-based OCR tool, and it is the right choice for private documents. Our free image to text tool runs the recognition engine locally in your browser using WebAssembly: drop in a JPG, PNG, screenshot, or iPhone photo, click Extract text, and copy the result or download it as a .txt file.
Because nothing is uploaded, it is safe for IDs, contracts, payslips, and medical documents, and there is no signup, watermark, or daily limit. It also tells you honestly when it struggles: uncertain words are highlighted instead of silently guessed, and if it detects a table it says so rather than handing you scrambled columns.
Method 2: Live Text on iPhone and Mac
If the image is on an Apple device, the feature is already built in. Open the photo (Photos app, Safari, or even the camera preview), touch and hold a word, and drag the selection handles: iOS treats printed text in images as selectable text. On a Mac, the same works in Preview and Photos with a normal click and drag.
Live Text is excellent for quick grabs like a Wi-Fi password on a router label or an address on a sign. Its limits: it works image by image, gives you no file output, and offers no help with tables.
Method 3: Google Lens on Android (and any browser)
Android's equivalent is Google Lens: open the photo in Google Photos, tap the Lens icon, and select the text to copy it. On desktop, images.google.com and the Chrome right-click menu ("Search image with Google Lens") do the same.
Lens recognizes text well, including many languages. The trade-off is privacy: the image is processed on Google's servers, so it is the wrong choice for sensitive documents.
Method 4: PowerToys Text Extractor on Windows
Windows users with Microsoft PowerToys installed can press Win+Shift+T, draw a rectangle over any text on screen, and the recognized text lands on the clipboard. It works on anything visible: images, videos on pause, apps that block copying. Recognition quality is decent on clean print, and everything stays on your machine. You need to install PowerToys once, and there is no file output, just the clipboard.
Method 5: The Google Docs trick
Upload an image to Google Drive, right-click it, and choose Open with Google Docs: Drive runs OCR and opens a document with the image on top and the recognized text underneath. Useful when you are already in Google Workspace and want the text in a document immediately. Same privacy caveat as Lens, and table structure is not preserved.
Method 6: When the image contains a table, use AI instead
Here is the fork in the road. Every method above outputs a stream of plain text, which is fine for paragraphs and fatal for tables. OCR reads line by line, so a photographed price list, bank statement, or lab report comes out with values detached from their columns, and you spend longer repairing the result than retyping it.
For tables, the right tool reads the image the way a person does and returns the structure, not just the characters. ScanPilot's image to Excel converter does exactly that: photograph the table, upload it, and download a real spreadsheet with rows, columns, and headers intact, and numbers stored as numbers. The first page converts free with no signup, so you can judge the result on your own document.
Handwriting: skip classic OCR entirely
The second case where plain OCR disappoints is handwriting. Classic engines were built for print; on a handwritten note they misread enough words to make the output untrustworthy. AI models read handwriting in context, which is why they get it right where OCR guesses.
If your image is a handwritten list, note, or filled-in form, go straight to an AI tool: ScanPilot's photo to text converter reads handwriting reliably and returns the text organized the way it appears on the page, copyable directly or downloadable as a file. For handwritten tables specifically, our guide on converting handwritten notes to Excel walks through the full workflow.
Quick reference
| Situation | Best method |
|---|---|
| Private document, any device | Free browser tool, nothing uploaded |
| Quick grab on iPhone or Mac | Live Text (built in) |
| Quick grab on Android | Google Lens |
| Text on screen in any app (Windows) | PowerToys Text Extractor (Win+Shift+T) |
| Already in Google Workspace | Google Docs OCR trick |
| Table in the image | AI image to Excel converter |
| Handwriting | AI photo to text converter |
For clean printed text, the free routes above cover you completely, and the browser tool keeps everything on your device. The moment structure or handwriting is involved, the AI converter is the difference between text you can use and text you have to fix.
Frequently Asked Questions
How do I extract text from an image for free?
Use a browser-based OCR tool: drop the image in, click extract, and copy the recognized text. Our free image to text tool does this without uploading your image anywhere, since recognition runs locally in your browser. Phone users can also use built-in features: Live Text on iPhone and Google Lens on Android both copy text straight from photos.
How do I copy text from a picture on my phone?
On iPhone, open the photo, touch and hold a word, and drag to select: that is Live Text, built into iOS. On Android, open the photo in Google Photos or Google Lens and tap the Lens icon, then select the text. Both let you copy printed text in seconds without installing anything.
Can I extract text from an image without uploading it?
Yes. Browser-based OCR can run entirely on your own device with WebAssembly, so the image never reaches a server. That matters for IDs, contracts, medical documents, and anything else you would not email to a stranger. Our free image to text tool works this way.
How do I get the text out of a photographed table?
Plain OCR reads tables line by line, so columns collapse into a jumble. For a table you want two things: the text and its structure. An AI converter reads the image the way a person does and rebuilds rows and columns, so a photographed price list or statement becomes a real spreadsheet instead of scrambled text.
Can OCR read handwriting?
Classic OCR engines are unreliable on handwriting: expect many wrong words. AI-based tools are a different story, since they read handwriting in context the way a person does. If your image is a handwritten note or a filled-in form, skip the classic tools and use an AI converter directly.
What makes an image OCR well?
Sharp text, decent light, and a straight-on angle. Shoot at the page's native size rather than zooming a small photo later, avoid shadows across the text, and prefer PNG for screenshots. If a photo OCRs badly, retaking it in better light fixes more than any setting.