PreP-OCR: A Complete Pipeline for Document Image Restoration and Enhanced OCR Accuracy

TL;DR

PreP-OCR is a comprehensive pipeline that combines document image restoration with semantic-aware post-OCR correction, significantly improving text extraction accuracy from degraded historical documents across multiple languages.

Contribution

It introduces a novel two-stage pipeline integrating synthetic data training for image restoration and language-aware post-correction, advancing OCR performance on historical documents.

Findings

Reduces character error rates by up to 70% on real historical documents.

Effectively handles multi-language historical texts.

Demonstrates the benefit of combining image restoration with linguistic correction.

Abstract

This paper introduces PreP-OCR, a two-stage pipeline that combines document image restoration with semantic-aware post-OCR correction to enhance both visual clarity and textual consistency, thereby improving text extraction from degraded historical documents. First, we synthesize document-image pairs from plaintext, rendering them with diverse fonts and layouts and then applying a randomly ordered set of degradation operations. An image restoration model is trained on this synthetic data, using multi-directional patch extraction and fusion to process large images. Second, a ByT5 post-OCR model, fine-tuned on synthetic historical text pairs, addresses remaining OCR errors. Detailed experiments on 13,831 pages of real historical documents in English, French, and Spanish show that the PreP-OCR pipeline reduces character error rates by 63.9-70.3% compared to OCR on raw images. Our pipeline demonstrates the potential of integrating image restoration with linguistic error correction for digitizing historical archives.