An Emacs minor feature that automatically inserts OCR (text recognition) output under images inserted via org-download, using Tesseract OCR. The OCR text is stored in a folded :ocr: drawer directly below the image link, making it searchable but unobtrusive.

Personally, I use this when taking org-roam notes so that screenshots are searchable.

• 🖼️ Automatically runs Tesseract OCR on images inserted via `org-download-clipboard`
• 📦 Inserts OCR text in an Org :ocr: drawer
• 🔍 OCR text is indexed by isearch and other Org search tools
• 🧹 The drawer is folded automatically to keep your buffer clean

• Emacs 27.1+
• org-download package
• Tesseract OCR installed and available on your PATH

Once submitted and accepted to MELPA:

(use-package org-ocr-drawer
  :ensure t
  :config
  (org-ocr-drawer-enable))

1. Clone this repository:

   git clone https://github.com/fkgruber/org-ocr-drawer
       

1. Add the directory to your load-path:

   (add-to-list 'load-path "/path/to/org-ocr-drawer")
   (require 'org-ocr-drawer)
   (org-ocr-drawer-enable)
       

1. Ensure Tesseract is installed (e.g. brew install tesseract on macOS).
2. In an Org buffer, run M-x org-download-clipboard to paste an image.
3. A drawer like this will be inserted:

   [[file:screenshot.png]]
   
   :ocr:
   Recognized text from the image...
   :end:
       

4. The :ocr: drawer will be folded automatically.

You can disable the OCR feature at any time with:

(org-ocr-drawer-disable)

This project is licensed under the MIT License.

Fred Gruber email {at} fredgruber {dot} org