How to Extract Text from an Image in Chrome (3 Ways)

OCR guide · 5 min read

Screenshots, scanned documents, product photos, and infographics often lock useful text inside pixels. When you cannot select the words with your cursor, you need optical character recognition (OCR) to read the image and turn it back into editable text. If you spend most of your day in Chrome, you do not have to leave the browser to do this.

This guide covers three practical ways to extract text from an image in Chrome, when each method works best, and how to get accurate results even from imperfect images.

Method 1: Right-click an image already on a web page

This is the fastest option when the image is already loaded in your browser: a diagram in an article, a screenshot someone shared, or a scanned page in an email preview.

Steps

1. Right-click directly on the image you want to read.
2. Choose an OCR option from the context menu. With a browser extension such as Textquill installed, you will see an "Extract text" entry.
3. Wait a moment for the recognition to finish, then copy the result or export it.

Because the extension already has access to the image on the page, there is no downloading, re-uploading, or hunting for the original file. Textquill runs the OCR on your device, so the picture is never sent to a server. That matters when the image is a receipt, an ID, or an internal document you would rather not upload anywhere.

Method 2: Select a region of the screen (area capture)

Sometimes the text is not a normal image at all. It might be inside a PDF viewer, a video frame, a slide, a chart label, or a chunk of a larger screenshot. In those cases you want to grab a specific rectangle of whatever is on screen rather than a single image file.

Steps

1. Press the capture hotkey. In Textquill the default is Alt+Shift+S.
2. Drag a box around just the text you need.
3. Release, and the selected area is read and converted to text.

Area select is the most flexible method because it does not care where the text lives. The main tip is to crop tightly: select only the words you want, not the whole window. A tighter selection means fewer stray graphics for the OCR engine to misread.

Method 3: Upload, drag, or paste an image file

When the image lives on your computer, phone, or clipboard rather than on a web page, you can feed it directly to an OCR tool.

• Upload: open the tool and pick the file from your device.
• Drag and drop: drag the image straight from a folder onto the extraction window.
• Paste: copy an image (for example with a screenshot shortcut) and paste it in with Ctrl+V or Cmd+V.

This method is handy for photos taken on a phone and synced to your desktop, or for a batch of scanned pages saved in a folder. Once the text is recognized, most tools let you copy it or export it as a TXT or Markdown file so you can keep formatting like headings and lists.

Tips for accurate OCR results

OCR quality depends heavily on the source image. A few adjustments make a large difference:

• Use the highest resolution you can. Tiny, blurry text is the top cause of errors. If you are capturing from a web page, zoom the page in first, then take the shot.
• Keep good contrast. Dark text on a light background reads best. Faded, low-contrast, or busy backgrounds trip up recognition.
• Straighten the text. Photos taken at an angle or rotated scans are harder to read. Crop and rotate so lines run horizontally.
• Pick the right language. Most tools default to English. If your text is in another script, choose the matching language so accented and non-Latin characters come through correctly. Textquill supports 16 languages for this reason.
• Avoid decorative fonts. Handwriting, script fonts, and heavy stylization are the least reliable. Clean, printed type gives the cleanest output.

Common problems and fixes

• Garbled or missing characters: usually a resolution or language issue. Recapture at a larger size or switch the recognition language.
• Only part of the text was caught: your selection likely cut off some lines. Redo the area select with a slightly wider box.
• A QR code or barcode instead of words: some tools, Textquill included, detect QR codes and return the encoded link or text directly rather than trying to read it as letters.
• Nothing happens on a protected page: browser or bank pages sometimes block extensions. Use the area-select or upload method on a screenshot instead.

Browser extension vs online OCR site vs copy-from-web

There are three broad categories of tools, and they suit different needs:

• Browser extensions live inside Chrome, so extracting text is a right-click or a hotkey away with no site to visit. On-device options like Textquill also keep the image on your machine, which is better for private documents and works offline.
• Online OCR sites require no install and can be convenient on a shared computer, but you upload your image to someone else's server. That is fine for public content, less ideal for anything sensitive, and it needs a connection.
• Copy-from-web is not OCR at all: if text is already selectable on a page, just highlight and copy it. Always try this first, because real text beats any recognition. OCR is for when the words are baked into an image.

For most people who work in Chrome, an on-device extension covers the everyday cases, an online site is a fallback when you cannot install anything, and plain copy-paste handles the many pages where the text was never an image to begin with.

FAQ