Low-Resource Heuristics for Bahnaric Optical Character Recognition Improvement

2026-01-06

arXiv.org

AI Summary

This study addresses the low OCR accuracy for endangered, low-resource languages such as Bahnar, which stems from severe data scarcity and document image degradation. To tackle this challenge, the authors propose an enhanced OCR framework specifically designed for such languages, integrating table/non-table region detection, image preprocessing, and a probability-driven heuristic post-processing strategy. This integrated approach effectively mitigates the adverse impact of poor document quality on recognition performance. Experimental results demonstrate that the proposed system significantly improves OCR accuracy on Bahnar documents, increasing it from 72.86% to 79.26%. The work thus offers a scalable and practical technical pathway for the digital preservation of endangered languages with limited linguistic resources.

Technology Category

• Natural Language Processing: Other
• Computer Vision: Language and Vision
• Knowledge Representation and Reasoning: Knowledge Representation Languages

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Bahnar, a minority language spoken across Vietnam, Cambodia, and Laos, faces significant preservation challenges due to limited research and data availability. This study addresses the critical need for accurate digitization of Bahnar language documents through optical character recognition (OCR) technology. Digitizing scanned paper documents poses significant challenges, as degraded image quality from broken or blurred areas introduces considerable OCR errors that compromise information retrieval systems. We propose a comprehensive approach combining advanced table and non-table detection techniques with probability-based post-processing heuristics to enhance recognition accuracy. Our method first applies detection algorithms to improve input data quality, then employs probabilistic error correction on OCR output. Experimental results indicate a substantial improvement, with recognition accuracy increasing from 72.86% to 79.26%. This work contributes valuable resources for Bahnar language preservation and provides a framework applicable to other minority language digitization efforts.

Problem

• Low-Resource
• Bahnaric
• Optical Character Recognition
• Document Digitization
• Minority Language Preservation

Innovation

• low-resource OCR
• probabilistic post-processing
• table detection
• minority language preservation
• Bahnaric script