Extract text from images using Tesseract OCR.
Use the DMS CLI:
dms plugins install ocrScannerOr manually:
git clone https://github.com/hthienloc/dms-ocr-scanner ~/.config/DankMaterialShell/plugins/ocrScanner- Screenshot OCR - Middle-click the icon to capture a screen region and scan it instantly
- Persistent History - Save and browse previous scans with images and text
- Clipboard scan - One-click to extract text from a clipboard image (right-click)
- Multi-language - English, Vietnamese, and 100+ other languages
- Auto-copy - Extracted text automatically copied to clipboard
- Save to file - Export as .txt
| Action | Result |
|---|---|
| Left click | Open scanner |
| Middle click | Scan from screenshot (region) |
| Right click | Scan from clipboard |
| Package | Installation |
|---|---|
tesseract |
sudo dnf install tesseract / sudo pacman -S tesseract |
tesseract-data-* |
Language packs (e.g., tesseract-langpack-eng) |
GPL-3.0
- Screenshot OCR: Capture and scan portions of the screen directly using the DMS screenshot tool.
- Persistent History: Implement a local database to store previous scans, including timestamps and source image references.
- On-device Translation: Integrate a lightweight translation service (or API) to translate extracted text instantly.
- Batch Processing: Support selecting multiple images for sequential OCR with a merged text output.
- Advanced Tesseract Options: Expose Page Segmentation Method (PSM) and OCR Engine Mode (OEM) settings in the UI for specialized documents.
- Image Preprocessing: Improve recognition accuracy by using ImageMagick to upscale small text, normalize contrast, and reduce noise before scanning.
| Language | Locale | Progress | Coverage | Status |
|---|---|---|---|---|
| Arabic | ar |
21/21 | 100.0% | 🟢 Complete |
| Bulgarian | bg |
21/21 | 100.0% | 🟢 Complete |
| German | de |
21/21 | 100.0% | 🟢 Complete |
| Esperanto | eo |
21/21 | 100.0% | 🟢 Complete |
| Spanish | es |
21/21 | 100.0% | 🟢 Complete |
| Persian | fa |
21/21 | 100.0% | 🟢 Complete |
| French | fr |
21/21 | 100.0% | 🟢 Complete |
| Hebrew | he |
21/21 | 100.0% | 🟢 Complete |
| Hungarian | hu |
21/21 | 100.0% | 🟢 Complete |
| Italian | it |
21/21 | 100.0% | 🟢 Complete |
| Japanese | ja |
21/21 | 100.0% | 🟢 Complete |
| Korean | ko |
21/21 | 100.0% | 🟢 Complete |
| Dutch | nl |
21/21 | 100.0% | 🟢 Complete |
| Polish | pl |
21/21 | 100.0% | 🟢 Complete |
| Portuguese | pt |
21/21 | 100.0% | 🟢 Complete |
| Russian | ru |
21/21 | 100.0% | 🟢 Complete |
| Swedish | sv |
21/21 | 100.0% | 🟢 Complete |
| Turkish | tr |
21/21 | 100.0% | 🟢 Complete |
| Ukrainian | uk |
21/21 | 100.0% | 🟢 Complete |
| Vietnamese | vi |
21/21 | 100.0% | 🟢 Complete |
| Chinese (Simplified) | zh_CN |
21/21 | 100.0% | 🟢 Complete |
| Chinese (Traditional) | zh_TW |
21/21 | 100.0% | 🟢 Complete |
