Optical character recognition meaning

Optical character recognition (OCR) is the process of converting scanned or photographed documents (for example, PDFs or TIFFs) into machine‑readable, searchable text for use in disclosure/discovery and document review. In practice, OCR allows legal teams to run keyword searches, deduplicate, apply analytics, and review and redact documents that originate as hard copy or image‑only files. In England and Wales, the term is expressly defined in cpr PD 31B para 5(9) as the computer‑facilitated recognition of printed or written text within an electronic image whose text content is otherwise not electronically searchable. While the CPR definition sits within e‑disclosure, the concept is used consistently across Scotland, Northern Ireland and Ireland in e‑discovery and disclosure workflows, typically as a descriptive technical term rather than a separately defined legal term. Key points: OCR accuracy depends on scan quality, language and handwriting; errors can affect search results, privilege review and redactions. Parties should record OCR settings and quality‑control steps, and consider proportionality and reasonableness when agreeing the scope of disclosure and reasonable searches. Common outputs include searchable PDFs and text files embedded or supplied with the images; OCR does not capture non‑textual content or metadata.