Feat: Add Tesseract OCR feature to selection tool

Tunahan Yardimcirequested to merge
tunahanyrd/okular:work-tunahan into master

Hello dear okular developers;

This merge request introduces a much-requested feature: Optical Character Recognition (OCR) for image selections.

This allows users to quickly extract unselectable text (from scanned documents or image-based PDFs) and copy it directly to the clipboard.

Implementation Details:

  • Tool: Uses Tesseract for OCR processing.

  • Workflow: A new "OCR to Clipboard" option is added to the context menu when an image area is selected in Selection Mode.

  • Image Quality Optimization (Crucial for Accuracy): To ensure high-quality recognition from low-DPI screen captures, the image selection is upscaled 3x and converted to Grayscale (Format_Grayscale8) before being passed to the Tesseract engine. This significantly boosts accuracy.

  • Localization: Tesseract is initialized with the multi-language setting "tur+eng" for broad utility.

  • Scope: This feature is implemented in okularpart and uses Tesseract_LIBRARIES for linking to maintain consistency with the project's existing CMake style.

    **Demonstration Video:** You can see the workflow in action here:

    Screencast_20251208_154635.webm

Merge request reports