OCR — optical character recognition — is the technology that reads text visible inside images: scanned documents, photographed signage, packaging, slides exported as pictures, screenshots. It converts pixels that happen to look like letters into actual text a system can index and search.

In a DAM, OCR closes a specific findability gap. A great deal of library content carries its most identifying information as pixels: the product name on a packshot label, the headline on a scanned print ad, the SKU on a datasheet saved as an image, the slogan on an event banner. None of that is findable by conventional search, because as far as the index is concerned an image contains no words. OCR extracts that text at ingestion and attaches it to the asset, so searching for a product name finds the packaging photo — not just the files someone remembered to tag with it.

OCR also feeds the rest of the pipeline. Extracted text becomes material for semantic search, gives automated rules something to match on, and rescues legacy archives — decades of scanned print material become searchable without anyone retyping a word. The practical caveat is that quality follows the source: clean scans and product labels read reliably, while low-resolution or heavily stylized text yields partial results, so OCR text is best treated as a searchable layer rather than a canonical transcription.

In freedam, OCR runs automatically during ingestion, and the extracted text is indexed for search alongside the asset's other metadata.

Related terms: Semantic Search, Metadata, Ingestion