Enhancing OCR Performance through Post-OCR Models: Adopting Glyph Embedding for Improved Correction

TL;DR

This paper presents a novel post-OCR correction model that incorporates glyph embeddings via CharBERT to improve correction accuracy, especially for correcting individual words, addressing limitations of traditional OCR systems.

Contribution

Introduces a new post-OCR correction approach using glyph embedding with CharBERT, enhancing correction performance over existing models.

Findings

Glyph embedding improves correction accuracy.

Post-OCR correction effectively addresses OCR deficiencies.

Model can correct individual words more accurately.

Abstract

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.