On the Use of Neural Text Generation for the Task of Optical Character Recognition

Georgios Theodoropoulos

10.1109/AICCSA47632.2019.9035333

2019-11

16th ACS/IEEE International Conference on Computer Systems and Applications AICCSA 2019

2161-5322

978-1-7281-5053-6

2019 IEEE/ACS 16TH INTERNATIONAL CONFERENCE ON COMPUTER SYSTEMS AND APPLICATIONS (AICCSA 2019)

2019-November

1-8

3-7 Nov. 2019

迪拜

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

Optical Character Recognition (OCR), is extraction of textual data from scanned text documents to facilitate their indexing, searching, editing and to reduce storage space. Although OCR systems have improved significantly in recent years, they still suffer in situations where the OCR output does not match the text in the original document. Deep learning models have contributed positively to many problems but their full potential to many other problems are yet to be explored. In this paper we propose a post-processing approach based on the application deep learning to improve the accuracy of OCR system (minimizing the error rate). We report on the use of neural network language models to accomplish the task of correcting incorrectly predicted characters/words by OCR systems. We applied our approach to the IAM handwriting database. Our proposed approach delivers significant accuracy improvement of 20.41% in F-score, 10.86% in character level comparison using Levenshtein distance and 20.69% in document level comparison over previously reported context based OCR empirical results of IAM handwriting database.

Neural text generation Optical character recognition OCR OCR post-processing language models neural language model text generation text prediction IAM database handwritten character recognition

其他

英语

CPCI-S ; EI

Computer Science

Computer Science, Information Systems ; Computer Science, Theory & Methods

WOS:000563487400108

20201508381020

Digital storage ; Database systems ; Computational linguistics ; Deep learning ; Learning systems

Ergonomics and Human Factors Engineering:461.4 ; Computer Theory, Includes Formal Logic, Automata Theory, Switching Theory, Programming Theory:721.1 ; Data Storage, Equipment and Techniques:722.1 ; Database Systems:723.3 ; Light/Optics:741.1

Web of Science

Georgios Theodoropoulos. On the Use of Neural Text Generation for the Task of Optical Character Recognition[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2019:1-8.