ImgPeel
ImgPeel · Peel OCR

Guide

ImgPeel (Peel OCR) turns screenshots, scans, and photos into editable text. Drop, pick, or paste an image — everything runs in your browser, and your pictures are never uploaded.

Good for everyday print text and mixed Chinese–English. Copy the result, or export TXT, Markdown, or Word. No sign-up, no usage limits, no watermarks.

What you can do

FeatureDetails
Add imagesDrag and drop, pick multiple files, or paste a screenshot (Ctrl/⌘V)
Batch processingUp to 20 images at once; switch between them anytime
Three modesFast, Balanced, or High accuracy — pick what fits the moment
Easy to reviewText line by line, with highlights on the original image
Copy allCopy the full result in one click
Export TXT / Markdown / WordSave one file, or a zip with one file per image for batches
Stays privateNo account, no limits, no watermark; images never leave your device

How to use

  1. Add your images

    Drop them in, click to choose (multi-select OK), or paste a screenshot with Ctrl/⌘V. Up to 20 at a time.

  2. Let it recognize

    Images are processed one by one with live progress. They never leave your device.

  3. Copy or export

    Check the text, copy everything, or export .txt / .md / .docx. For several images, download a zip with one file each.

Which mode should I pick?

Faster modes download less and finish sooner; higher accuracy needs more space. Switch anytime — once a mode is downloaded, it stays ready for next time.

Fast

PP-OCRv6 TinyAbout 6 MB

Lightest and quickest — fine for screenshots and clear documents; a good pick on mobile data.

Balanced

PP-OCRv6 SmallAbout 30 MB

A solid everyday choice — steadier on smaller text and busier layouts.

High accuracy

PP-OCRv6 MediumAbout 132 MB

Most careful results; first download is larger — best on a computer with Wi‑Fi.

FAQ

Are my images uploaded to a server?
No. Recognition runs in your browser. Your pictures stay on this page and are not sent to our servers. You can even turn off Wi‑Fi after the first setup and keep recognizing.
Which languages and image formats are supported?
Chinese, English, numbers, and common symbols — mixed text needs no language switch. Images: PNG, JPG, WebP. Add them by drop, file picker, or Ctrl/⌘V paste.
How do I choose among the three modes?
Fast is lightest and quickest — great for screenshots and clear docs. Balanced is a good everyday default. High accuracy is the most careful, but the first download is larger — use Wi‑Fi on a computer when you can.
Why does the first visit take a while?
The first time needs a one-time download so recognition can run offline in your browser. After that it opens much faster from local storage.
Can I use it on a phone?
Yes — take a photo or pick from your album. On mobile data, Fast or Balanced is usually more comfortable; High accuracy is heavier.
How do I export results?
Copy all text, or download .txt / .md / .docx for the current image. After a batch, use the zip options to get one file per image.
Do I need an account or payment?
No. Open the page and use it — no login, quotas, or watermarks.
What if recognition is inaccurate?
Clear, upright text with good contrast works best. Handwriting, fancy fonts, blurry or tiny text are harder — try High accuracy, or retake a sharper photo.