JaPOC: Japanese Post-OCR Correction Benchmark using Vouchers

TL;DR

This paper introduces a new benchmark for Japanese post-OCR correction on vouchers, evaluates existing methods, and proposes simple language model-based baselines that significantly improve recognition accuracy.

Contribution

It creates the first publicly available OCR error correction benchmark for Japanese vouchers and demonstrates the effectiveness of simple language model baselines.

Findings

Error correction significantly improves OCR accuracy on vouchers.

Proposed baselines outperform existing methods.

Benchmark enables future research in Japanese OCR correction.

Abstract

In this paper, we create benchmarks and assess the effectiveness of error correction methods for Japanese vouchers in OCR (Optical Character Recognition) systems. It is essential for automation processing to correctly recognize scanned voucher text, such as the company name on invoices. However, perfect recognition is complex due to the noise, such as stamps. Therefore, it is crucial to correctly rectify erroneous OCR results. However, no publicly available OCR error correction benchmarks for Japanese exist, and methods have not been adequately researched. In this study, we measured text recognition accuracy by existing services on Japanese vouchers and developed a post-OCR correction benchmark. Then, we proposed simple baselines for error correction using language models and verified whether the proposed method could effectively correct these errors. In the experiments, the proposed error correction algorithm significantly improved overall recognition accuracy.