Survey of Post-OCR Processing Approaches

Optical character recognition (OCR) is one of the most popular techniques used for converting printed documents into machine-readable ones. While OCR engines can do well with modern text, their performance unfortunately significantly reduced on historical materials. Additionally, many texts have already been processed by various out-of-date digitisation techniques. As a consequence, digitised are noisy and need to be post-corrected. This article clarifies importance enhancing quality results studying effects information retrieval natural language processing applications. We then define post-OCR problem, illustrate its typical pipeline, review state-of-the-art approaches. Evaluation metrics, accessible datasets, resources, useful toolkits also reported. Furthermore, work identifies current trend outlines some research directions this field.