OCR Error Correction Using Character Correction and Feature-Based Word Classification

Abstract

This paper explores the use of a learned classifier for post-OCR text correction. Experiments with the Arabic language show that this approach, which integrates a weighted confusion matrix and a shallow language model, improves the vast majority of segmentation and recognition errors, the most frequent types of error on our dataset.

Proceedings of the 12th IAPR International Workshop on Document Analysis Systems (DAS2016), Santorini, Greece, April 11-14, 2016

Cited by in corpus (1)

• Toward the Optimized Crowdsourcing Strategy for OCR Post-Correction