Optical Character Recognition (OCR) Services

The State Library is seeking to identify suitably qualified and experienced companies to provide optical character recognition (OCR) services for digitised from the Library’s collection. As one part of the State Library’s assessment process, respondents will be given a test batch of digitised items (the SOURCE IMAGES). The Respondent will be required to OCR all files in the test batch, deliver specified cropped SCREEN JPEG files, METS / ALTO and PDF/A files, as well as any administrative and technical metadata requested. Annexure F sets out the instructions for “OCR Pilot Test”. The OCR Services (the “Deliverables”) required by the Libray include: The creation of cropped and de-skewed SCREEN JPEG’s from SOURCE IMAGES that conform to the specifications provided by the SLNSW The creation of OCR from SCREEN JPEGs The creation of METS, Rosetta METS and ALTO that conforms to specified schemas and reference the SCREEN JPEGs Addition of specific metadata to the Rosetta METS, METS and ALTO files The creation of searchable PDF/A for each book/serial/item that conforms to the standards provided by the SLNSW for specific administrative metadata Quality Assurance of the Deliverables produced from the services described in Section 5 of this document File transfer of the Deliverables back to SLNSW Retention of a copy of the digitised files until the integrity and quality acceptance of the transferred files has been confirmed by SLNSW Project management for the term of the contract

Comments are closed.