Domain Adaptation for Document Image Binarization via Domain Classification
Por favor, use este identificador para citar o enlazar este ítem:
http://hdl.handle.net/10045/121474
Título: | Domain Adaptation for Document Image Binarization via Domain Classification |
---|---|
Autor/es: | Garrido Muñoz, Carlos | Sánchez Hernández, Adrián | Castellanos, Francisco J. | Calvo-Zaragoza, Jorge |
Grupo/s de investigación o GITE: | Reconocimiento de Formas e Inteligencia Artificial |
Centro, Departamento o Servicio: | Universidad de Alicante. Departamento de Lenguajes y Sistemas Informáticos |
Palabras clave: | Document Image Binarization | Deep Neural Networks | Unsupervised Domain Adaptation | Domain Classifier |
Área/s de conocimiento: | Lenguajes y Sistemas Informáticos |
Fecha de publicación: | 2021 |
Editor: | IOS Press |
Cita bibliográfica: | Garrido-Munoz, Carlos, et al. “Domain Adaptation for Document Image Binarization via Domain Classification”. In: Tallón-Ballesteros, Antonio J. (Ed.). Modern Management based on Big Data II and Machine Learning and Intelligent Systems III. Proceedings of MMBD 2021 and MLIS 2021. Amsterdam: IOS Press BV, 2021. ISBN 978-1-64368-224-2, pp. 569-582 |
Resumen: | Binarization represents a key role in many document image analysis workflows. The current state of the art considers the use of supervised learning, and specifically deep neural networks. However, it is very difficult for the same model to work successfully in a number of document styles, since the set of potential domains is very heterogeneous. We study a multi-source domain adaptation strategy for binarization. Within this scenario, we look into a novel hypothesis where a specialized binarization model must be selected to be used over a target domain, instead of a single model that tries to generalize across multiple domains. The problem then boils down to, given several specialized models and a new target set, deciding which model to use. We propose here a simple way to address this question by using a domain classifier, that estimates which of the source models must be considered to binarize the new target domain. Our experiments on several datasets, including different text styles and music scores, show that our initial hypothesis is quite promising, yet the way to deal with the decision of which model to use still shows great room for improvement. |
Patrocinador/es: | This paper has been supported by Generalitat Valenciana through grant ACIF/2019/042 and project GV/2020/030, and Universidad de Alicante through project GRE19-04. The first two authors carried out this work as recipients of a grant from the Office for Educational Quality and Innovation of the University of Alicante, within the collaboration agreement with Banco de Santander S.A. |
URI: | http://hdl.handle.net/10045/121474 |
ISBN: | 978-1-64368-224-2 | 978-1-64368-225-9 |
DOI: | 10.3233/FAIA210289 |
Idioma: | eng |
Tipo: | info:eu-repo/semantics/bookPart |
Derechos: | © 2021 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). |
Revisión científica: | si |
Versión del editor: | https://doi.org/10.3233/FAIA210289 |
Aparece en las colecciones: | INV - GRFIA - Capítulos de Libros |
Archivos en este ítem:
Archivo | Descripción | Tamaño | Formato | |
---|---|---|---|---|
Domain-Adaptation-for-Document-Image-Binarization-via-Domain-Classification.pdf | 1,05 MB | Adobe PDF | Abrir Vista previa | |
Este ítem está licenciado bajo Licencia Creative Commons