Enhancing OCR Performance through Post-OCR Models: Adopting Glyph Embedding for Improved Correction
This addresses OCR accuracy problems for document digitization users, but appears incremental as it builds on existing post-OCR correction approaches.
The study tackled OCR limitations by developing a post-OCR correction model that incorporates glyph embedding to capture visual character characteristics, achieving superior results including individual word correction.
The study investigates the potential of post-OCR models to overcome limitations in OCR models and explores the impact of incorporating glyph embedding on post-OCR correction performance. In this study, we have developed our own post-OCR correction model. The novelty of our approach lies in embedding the OCR output using CharBERT and our unique embedding technique, capturing the visual characteristics of characters. Our findings show that post-OCR correction effectively addresses deficiencies in inferior OCR models, and glyph embedding enables the model to achieve superior results, including the ability to correct individual words.