Detection Masking for Improved OCR on Noisy Documents
Abstract
Optical Character Recognition (OCR), the task of extracting textual information from scanned documents is a vital and broadly used technology for digitizing and indexing physical documents. Existing technologies perform well for clean documents, but when the document is visually degraded, or when there are non-textual elements, OCR quality can be greatly impacted, specifically due to erroneous detections. In this paper we present an improved detection network with a masking system to improve the quality of OCR performed on documents. By filtering non-textual elements from the image we can utilize document-level OCR to incorporate contextual information to improve OCR results. We perform a unified evaluation on a publicly available dataset demonstrating the usefulness and broad applicability of our method. Additionally, we present and make publicly available our synthetic dataset with a unique hard-negative component specifically tuned to improve detection results, and evaluate the benefits that can be gained from its usage
Cite
@article{arxiv.2205.08257,
title = {Detection Masking for Improved OCR on Noisy Documents},
author = {Daniel Rotman and Ophir Azulai and Inbar Shapira and Yevgeny Burshtein and Udi Barzelay},
journal= {arXiv preprint arXiv:2205.08257},
year = {2022}
}