Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Confidence-Aware Document OCR Error Detection

Arthur Hemmer, Mickaël Coustaty, Nicola Bartolo, Jean-Marc Ogier

TL;DR

ConfBERT is developed, a BERT-based model that incorporates OCR confidence scores into token embeddings and offers an optional pre-training phase for noise adjustment, demonstrating that integrating OCR confidence scores can enhance error detection capabilities.

Abstract

Optical Character Recognition (OCR) continues to face accuracy challenges that impact subsequent applications. To address these errors, we explore the utility of OCR confidence scores for enhancing post-OCR error detection. Our study involves analyzing the correlation between confidence scores and error rates across different OCR systems. We develop ConfBERT, a BERT-based model that incorporates OCR confidence scores into token embeddings and offers an optional pre-training phase for noise adjustment. Our experimental results demonstrate that integrating OCR confidence scores can enhance error detection capabilities. This work underscores the importance of OCR confidence scores in improving detection accuracy and reveals substantial disparities in performance between commercial and open-source OCR technologies.

Confidence-Aware Document OCR Error Detection

TL;DR

ConfBERT is developed, a BERT-based model that incorporates OCR confidence scores into token embeddings and offers an optional pre-training phase for noise adjustment, demonstrating that integrating OCR confidence scores can enhance error detection capabilities.

Abstract

Optical Character Recognition (OCR) continues to face accuracy challenges that impact subsequent applications. To address these errors, we explore the utility of OCR confidence scores for enhancing post-OCR error detection. Our study involves analyzing the correlation between confidence scores and error rates across different OCR systems. We develop ConfBERT, a BERT-based model that incorporates OCR confidence scores into token embeddings and offers an optional pre-training phase for noise adjustment. Our experimental results demonstrate that integrating OCR confidence scores can enhance error detection capabilities. This work underscores the importance of OCR confidence scores in improving detection accuracy and reveals substantial disparities in performance between commercial and open-source OCR technologies.
Paper Structure (14 sections, 5 equations, 2 figures, 3 tables)

This paper contains 14 sections, 5 equations, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Data Acquisition
  4. Datasets
  5. OCR
  6. OCR Statistics
  7. Confidence-Aware Error Detection
  8. Architecture
  9. Additional Pre-training
  10. Experiments
  11. Impact of $\alpha$
  12. Limitations
  13. Conclusion
  14. Acknowledgments.

Figures (2)

  • Figure 1: OCR and ground truth (GT) bounding box alignment strategy. First a corresponding GT box is found for each OCR box and vice-versa (left), the two mappings are merged and the connected components should contain the same information (right).
  • Figure 2: Relative improvement in $F_1$ for different values of $\alpha$ compared to $\alpha = 0$.