Efficient and effective OCR engine training

Clausner, C ORCID: https://orcid.org/0000-0001-6041-1002, Antonacopoulos, A ORCID: https://orcid.org/0000-0001-9552-0233 and Pletschacher, S 2019, 'Efficient and effective OCR engine training' , International Journal on Document Analysis and Recognition . (In Press)

[img]
Preview
PDF - Published Version
Available under License Creative Commons Attribution 4.0.

Download (1MB) | Preview

Abstract

We present an efficient and effective approach to train OCR engines using the Aletheia document analysis system. All components required for training are seamlessly integrated into Aletheia: training data preparation, the OCR engine’s training processes themselves, text recognition, and quantitative evaluation of the trained engine. Such a comprehensive training and evaluation system, guided through a GUI, allows for iterative incremental training to achieve best results. The widely used Tesseract OCR engine is used as a case study to demonstrate the efficiency and effectiveness of the proposed approach. Experimental results are presented validating the training approach with two different historical datasets, representative of recent significant digitisation projects. The impact of different training strategies and training data requirements is presented in detail.

Item Type: Article
Schools: Schools > School of Computing, Science and Engineering
Journal or Publication Title: International Journal on Document Analysis and Recognition
Publisher: Springer Verlag
ISSN: 1433-2833
Related URLs:
Depositing User: Mr Christian Clausner
Date Deposited: 14 Oct 2019 11:03
Last Modified: 01 Nov 2019 16:30
URI: http://usir.salford.ac.uk/id/eprint/52696

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year