RoundTripOCR: A Data Generation Technique for Enhancing Post-OCR Error Correction in Low-Resource Devanagari Languages

TL;DR

This paper introduces RoundTripOCR, a synthetic data generation method for low-resource Devanagari languages, improving post-OCR error correction by framing errors as translation problems and leveraging transformer models.

Contribution

It presents a novel data generation technique and a transformer-based approach for OCR error correction tailored to low-resource Devanagari languages, with new datasets released.

Findings

Enhanced OCR error correction accuracy demonstrated

Effective use of translation techniques for error correction

New datasets for Hindi, Marathi, Bodo, Nepali, Konkani, Sanskrit

Abstract

Optical Character Recognition (OCR) technology has revolutionized the digitization of printed text, enabling efficient data extraction and analysis across various domains. Just like Machine Translation systems, OCR systems are prone to errors. In this work, we address the challenge of data generation and post-OCR error correction, specifically for low-resource languages. We propose an approach for synthetic data generation for Devanagari languages, RoundTripOCR, that tackles the scarcity of the post-OCR Error Correction datasets for low-resource languages. We release post-OCR text correction datasets for Hindi, Marathi, Bodo, Nepali, Konkani and Sanskrit. We also present a novel approach for OCR error correction by leveraging techniques from machine translation. Our method involves translating erroneous OCR output into a corrected form by treating the OCR errors as mistranslations in a parallel text corpus, employing pre-trained transformer models to learn the mapping from erroneous to correct text pairs, effectively correcting OCR errors.