How to Extract Text from an Image on Windows

An image stores colored pixels, not letters. To your PC, the words in a screenshot or in a photo of a page are shapes like any other, so there is nothing to select or copy until something recognizes them.

That recognition step is OCR. It looks at the shapes, works out which letters they are and writes them out as plain text, which you can then paste into an email, a spreadsheet or a document.

Yes, one image at a time. In Windows 11, press Windows logo key + Shift + S, capture the image on screen, open the capture in Snipping Tool, click Text actions and copy the text. PowerToys, a free Microsoft app you install separately, adds Text Extractor, which copies text from anything on screen.

OneNote can do it as well: paste a picture into a page, right-click it and copy the text from the picture.

These are handy for one phone number or one paragraph. They stop being handy with twenty receipts or a folder of screenshots: every image is captured, recognized and pasted by hand, and nothing is saved as a file.

File Acture's Extract Text reads the text in photos, scans and screenshots with Windows' own text recognition, on your PC. Drop one image or a whole batch, and each one comes back as a text file named after it, like receipt-text.txt.

Under Options, Multiple images decides what a batch becomes: Separate text file for each image, or Combine into one text file, where each image's text sits under its file name (the file is called Extracted Text unless you change File name under Result). Copy text to clipboard is on by default, so the text is ready to paste as soon as the run ends.

The text files are saved to Documents › File Acture › Images unless you choose another folder, and your images stay as they are.

Text recognition reads what it can see, so a sharp, straight image gives the cleanest text.

• Pick the language. Choose the text's language under Document Language, especially for accents and non-Latin scripts.
• Use sharp, straight images. Blurry, tilted or tiny text is easy to misread. Crop to the text or retake the photo if you can.
• Expect plain text. Lines come back in reading order, but tables, columns and fonts are not kept; check numbers before you rely on them.
• Handwriting is hit and miss. Printed text reads well; handwriting often does not.
• Need a searchable PDF instead? Put the photos into one PDF with Images to PDF, then run OCR PDF: the pages keep their look and the text becomes searchable.

Online converters need your image on their servers, and screenshots and photos of documents often hold more than you meant to share: account numbers, addresses, names in the background. File Acture recognizes the text on your PC, so the images never leave it; the free version needs an internet connection only to count its runs.

The free version includes 3 runs on each PC, with up to 5 images per run. For bigger batches, see pricing.