Improving Accessibility in Interlibrary Loan Using OCR

Abstract Optical character recognition (OCR) is used to identify typed, printed and handwritten text, and encode it in a digital format. OCR has been used in libraries since the 1990s, but only recently have libraries started using it in interlibrary loan (ILL). This study focused on seeing if OCR could be applied to scans made in an ILL unit and sent to patrons while incurring minimal costs and requiring few new technical skills. Copies of 20 articles requested through ILL were used to test the process and accuracy of Adobe Acrobat Pro DC in creating searchable PDFs. The process involved calculating the accuracy of automatic OCR results, accuracy after manual corrections were made, and noting any problems that arose. The results showed that searchable PDFs could be created using Adobe Acrobat Pro DC that were 95% accurate or more in 12 of the 20 test files. Adobe Acrobat Pro DC had difficulties processing some information and text, but this provides a good starting point for an ILL unit desiring to provide their patrons with accessible materials.