Please use this identifier to cite or link to this item:
https://scholarhub.balamand.edu.lb/handle/uob/7160
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chammas, Michel | en_US |
dc.contributor.author | Makhoul, Abdallah | en_US |
dc.contributor.author | Demerjian, Jacques | en_US |
dc.contributor.author | Dannaoui, Elie | en_US |
dc.date.accessioned | 2024-01-12T07:12:33Z | - |
dc.date.available | 2024-01-12T07:12:33Z | - |
dc.date.issued | 2024-12-06 | - |
dc.identifier.issn | 13807501 | - |
dc.identifier.uri | https://scholarhub.balamand.edu.lb/handle/uob/7160 | - |
dc.description.abstract | The extraction of paleographical features is an important task to study the identity of the text in the Historical Manuscripts. One of the major features is the identification of the writer or copyist. Many researchers have worked on an automated system for writer identification, and with the development of deep learning techniques many approaches have been proposed. Most of the previous studies have developed a multi-steps system, while very few of them performed an End-to-End approach. Most of the systems rely on a pre-processing step to prepare the data in order to facilitate recognition. This paper presents an End-to-End deep learning system for writer identification, tested on four different datasets: ICDAR19 and ICFHR20 (Latin datasets), KHATT and Balamand (Arabic datasets). The system is based on the Deep-TEN approach using a customized ResNet-50 network for features and local descriptor extraction with an integration of a NetVLAD end-layer to compute and encode the global descriptor. It was compared with our state-of-the-art system, winner of ICFHR20 HisFrag competition, and showed an interesting performance on all datasets without any pre-processing techniques. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | Springer | en_US |
dc.subject | Writer identification | en_US |
dc.subject | historical documents | en_US |
dc.subject | Deep learning | en_US |
dc.subject | Document Analysis | en_US |
dc.subject | End-to-end | en_US |
dc.subject | Arabic manuscripts | en_US |
dc.title | An End-to-End deep learning system for writer identification in handwritten Arabic manuscripts | en_US |
dc.type | Journal Article | en_US |
dc.identifier.doi | https://doi.org/10.1007/s11042-023-17303-8 | - |
dc.identifier.scopus | 2-s2.0-85178879423 | - |
dc.identifier.url | https://api.elsevier.com/content/abstract/scopus_id/85178879423 | - |
dc.contributor.affiliation | Department of Computer Science | en_US |
dc.contributor.affiliation | Institute of History Archeology and Near Eastern Studies | en_US |
dc.description.volume | 83 | en_US |
dc.description.issue | 18 | en_US |
dc.description.startpage | 54569 | en_US |
dc.description.endpage | 54589 | en_US |
dc.date.catalogued | 2024-06-27 | - |
dc.description.status | Published | en_US |
dc.identifier.openURL | https://link.springer.com/article/10.1007/s11042-023-17303-8 | en_US |
dc.relation.ispartoftext | Multimedia Tools and Applications | en_US |
crisitem.author.parentorg | Faculty of Arts and Sciences | - |
Appears in Collections: | Institute of History Archeology and Near Eastern Studies |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.