Find Text & Recognize Text (OCR)

Find Text & Recognize Text (OCR)

LocusBIM has two complementary text features for PDFs: Find Bar (Ctrl+F) for cross-page text search, and Recognize Text (OCR) for extracting searchable text from scanned or image-only drawings.

Find Bar (Ctrl+F)

Press Ctrl+F (or Cmd+F on macOS) anywhere in the PDF view to open the Find Bar. Type your query. Search runs as you type and scans every page of the document. The counter shows "X of Y" matches; while a long document is still being scanned it shows "Scanning N/M…" so you know it isn't stuck.

Matches are highlighted in yellow on the current page, with the active match in orange. Navigating to a match on a different page automatically switches to that page. Matches from native PDF text and OCR'd text are merged transparently no separate search modes.

The Find Bar is a real toolbar: drag it anywhere on screen, toggle it from View → Find Bar, and its position persists across sessions like every other dockable toolbar.

Recognize Text (OCR) for scanned PDFs

When a PDF page is image-only (a scanned drawing, photo of a redline, or a vector PDF with the text layer stripped), pdfium has nothing to give Find Bar and text selection doesn't work. LocusBIM detects this on doc-open and shows a yellow "This page is a scan" banner above the page.

Click Run OCR in the banner, or open the OCR dialog from the Document toolbar → Recognize Text (OCR) button. The dialog offers two choices:

1. Current page: fast (~5–10 s on a typical sheet) for testing
2. All pages: runs sequentially with a live progress bar showing "page N of M (recognizing text, X%)"

Recognised text is saved to the local database so reopening the document doesn't re-incur the OCR cost. Once a page is OCR'd, its banner disappears, text becomes selectable, and Find Bar finds matches in it.

Running OCR again skips pages that already have recognized text, so a re-run only processes new pages and pages where nothing was recognized last time. The dialog shows how many pages will be skipped; turn off Skip pages that already have recognized text to force a full re-run, for example after switching the OCR language. Rotating or cropping a page clears its saved text automatically, so the next run picks that page up again.

Language Support: English + French

LocusBIM ships with bundled OCR language packs for English and French. The OCR dialog has a language picker that defaults to English; pick Français for drawings whose annotations are in French (common on Quebec projects + dual-language federal work). Both languages run on the same Tesseract.js engine; there is no quality or speed difference between them.

Plan availability

Find Bar and Recognize Text (OCR) are both included in every paid plan. During the 30-day trial they're available too, with no feature restrictions.

Automatic OCR on open (optional)

In Preferences → General, enable "Run Recognize Text automatically on scanned PDFs" to skip the manual step entirely. When LocusBIM detects an image-only document, OCR starts in the background across every page. Disabled by default, the same default Adobe Acrobat uses.