Extract text from image

This article is the saved-picture job. The file already lives on this PC: a png in Downloads, a jpg from a phone, a screenshot you stored, a receipt photo, a slide export. You want the words out of that file, into the clipboard or into a text file you keep. Drawing a box around words that are only on screen right now sits on OCR. The short install sits home.

This is Text-Grab. MIT. Joseph Finney publishes it. We did not make it. Official files live on GitHub Releases. The repo is TheJoeFin/Text-Grab. This site does not host the installer. Skip Softonic, CNET, and unofficial wrappers.

The button on this page goes to Install on the home page. Trust notes sit on Image to text download safe.

Install so you can extract text from an image on Windows

There is no winget command. Do not paste a made-up package id. Copy this filename:

Text-Grab.exe

Open the latest GitHub release. For most PCs take Text-Grab.exe. On ARM Windows take Text-Grab-arm64.exe. Self-contained zips also sit on that release if you want a zip. Live tag observed: V4.15.0, 1 August 2026. Do not invent a newer number here. File notes sit on Image to text releases. LICENSE sits on Image to text GitHub.

Run the exe on the same machine. Files stay on GitHub. If SmartScreen warns on a first run, read the publisher. Prefer the GitHub asset over a renamed copy from search.

How to extract text from an image

1. Open GitHub Releases, pick Text-Grab.exe for x64 or Text-Grab-arm64.exe for ARM, and run it. There is no winget command.
2. Open Text-Grab. You should get a window, not a website that asks you to upload the picture.
3. Open a picture you already have on this PC. A supported image path can open in Grab Frame. Point at the file you saved, not at a tab you never stored.
4. Run recognition on that picture. Wait until the words appear as text you can select.
5. Copy the words to the clipboard, or keep them in Edit Text Window and save a file. Check a line you already know before you trust the whole dump.

That list is the extract text from image path. You are reading a file that already lives here. You are not grabbing a live screen region as the whole job. That region job sits on OCR.

What extract text from image is on this PC

A picture can hold letters and still refuse copy. The letters are paint. Extract text from image turns that paint into text you can paste or save. Text-Grab is one MIT program that does it in a desktop window on Windows 10 or later. You install it, you open the file, you run recognition, you copy or you save.

This page stays on a file you already have. Stay here when the picture is on disk and you want the words out of it. If you only needed to lasso words that are currently on a monitor, that sits on OCR.

The app does not create the picture. If the file is missing, still in a chat you never stored, or still named .crdownload, the window has nothing honest to read. Save the picture first. Then open that path.

Open a picture you already have

Start with one file you can recognize. A screenshot of a setting, a photo of a label, a slide export. Confirm the app shows words. Then open the next file. A first run that has never finished one good picture is a bad place to dump a whole camera roll.

File open in the app is the usual path. Passing a supported image path into Grab Frame also works. We will not invent extra menu names. If the picker cannot see the file, you are in the wrong folder. Browse to where you actually saved it.

A picture on a USB stick needs the stick plugged in. Ejecting the stick while recognition is running can fail the read. Copy the image to Documents if you need the words after the stick leaves. A picture on a network share is still a file. Wait until the share is available.

png and jpg are the usual types people already have. A supported PDF can also open in Grab Frame in this project. This article still stays on a file you saved, not on printing a new scan. If the name is not a finished image type, wait until it is.

Clipboard or a file you keep

Clipboard is enough when you will paste once into mail, a form, or a document. Copy. Paste. You are done. The picture stays on disk. Copying does not delete it.

Save a text file when you need the words later, or when the dump is long. Edit Text Window is the place to keep the result, clean spaces, and write it out. We will not invent a cloud folder for this project. You pick the destination on this PC.

If you only needed one line from a busy poster, copy that line. You do not have to save a full transcript of every pixel. Check the line against the picture. If a digit is wrong, fix it in the editor. Recognition is a draft. You still own the file you save.

First run

After you run the exe, open the app once before you hunt a website named extract. Open one picture you trust. Confirm words appear as text. If the window is empty, you have not opened a file yet, or the file has no letters, or the path failed.

Windows may show a SmartScreen prompt on a first exe. Read the publisher. Prefer GitHub Releases. The exe path is the honest one on this site. Do not fetch a second setup from search.

Updates live on GitHub Releases. We do not invent a newer number than V4.15.0 here. Read that page. Usual Windows pick is still Text-Grab.exe. The arm64 exe is still for ARM Windows. The self-contained zip is still a zip of this project.

When the picture has no usable words

A photo of a cat will not become a paragraph. A zero-byte file will not show letters. Download or save it again from the place you got it. A file still named .tmp is not finished. Wait until the name is a real image type.

A dark photo of a whiteboard, shot from an angle, will miss. Retake it, or crop to the line you need in any viewer you already have, save that crop, then open the crop. This page will not straighten a lens. It will not host a recovery tool.

A screenshot of a screenshot stacked several times can fail. Use the original save if you still have it. A tiny thumbnail emailed as a preview may be too small. Ask for the full picture. Then extract again.

Windows OCR languages follow language packs in Windows Settings. If the picture is a script Windows has not added, you can get empty output or the wrong letters. Add the language pack on the PC, then open the same file again. We do not invent a language list here. Check Windows.

A folder of pictures

You are not limited to one file. This project can process a folder of images and collect the output in Edit Text Window. Use that when you already saved a set of receipts, labels, or screenshots and you want the words in one place. Point at the folder you control. Do not point at a library you do not own.

Start with a small folder you can check. Confirm one file came out readable. Then run the rest. A bulk pass that has never been checked is a bad place to archive. You can write .txt output for processed items. You still read a sample. Garbage in a whole folder is still garbage, just more of it.

Command-line flags exist in the project if you want a path without a window. --windowless with a file path reads or recognizes and copies without opening a window. A folder path can open Edit Text Window and scan images in that directory. We will not invent extra flags. If you only needed one picture, ignore the folder path and open that file.

School PCs, labs, and uninstall

A lab image may block unknown exe files. Then this extract text from image install is closed for that hour. Do not fetch a wrapper to get around policy. Use what is already on the image, or wait until you control the PC.

A borrowed laptop you return in twenty minutes is a bad place to leave Text-Grab.exe. Use what is already on the machine, or do not install. Do not leave someone else a folder of other people's receipts.

Uninstall is Apps in Windows if the build registered there, or delete the unpacked zip if you used a zip. Quitting the window is not uninstall. If you only wanted one picture, quit first. If you want the program gone, remove it. Fetch again from GitHub Releases, not from a wrapper.

x64 exe, arm64 exe, and zip

Usual Windows pick is Text-Grab.exe. ARM Windows uses Text-Grab-arm64.exe. The same GitHub release also lists self-contained zips. They are different packages of the same project. Pick one path. Do not run the exe and unpack the zip on top of each other because the first one felt slow.

Match the file name on the release. Do not invent a newer tag than V4.15.0. Hash notes, if any, live on GitHub Releases, not here. The install pointer is Install on the home page.

What this extract text from image job will not do

It will not be the live-screen grab. That sits on OCR. It will not upload the picture to a website. It will not host the installer.

It will not be a scanner. It will not take a new photo for you. It will not be a star rating. It will not invent a winget id. Files stay on GitHub.

It will not fix a picture that never had readable letters. It will not wait on a wrapper named setup. Skip Softonic, CNET, and unofficial wrappers.

Shared machines and the next file

On a family PC, say if you installed a new program that reads pictures. On a work PC, follow local policy. If policy blocks unknown exe files, stop. Softonic is still the wrong host in that hour.

Open the next picture the same way. Install once. Extract many files. You do not reinstall for each image. If a new file fails, the file is the suspect, not a missing second program.

How do I extract text from an image?

Take Text-Grab.exe from GitHub Releases, or the arm64 exe. Run it. Open a picture you already have. Run recognition. Copy or save. There is no winget command. The install pointer is Install on the home page.

Where does the text go after I extract it?

The clipboard, if you copied. A text file, if you saved from Edit Text Window. The picture stays where you opened it unless you moved it. Closing the app does not delete the image.

Why is the extracted text wrong?

The picture may be small, dark, angled, or a script Windows has not added. Open a sharper save. Add the language pack if needed. Fix the line in Edit Text Window before you keep it. Recognition is a draft of the letters in the file.

Did you make this extract text from image?

No. Joseph Finney publishes Text-Grab under MIT. Official files live on GitHub Releases. This site only points at that page. We do not host the installer. Back home if you want the short version.