Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Neural network language models for off-line handwriting recognition

  • Zamora Martínez, Francisco Julián
  • Frinken, V.
  • Fischer, A.
  • Bunke, Horst
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
[EN] Unconstrained off-line continuous handwritten text recognition is a very challenging task which has been recently addressed by different promising techniques. This work presents our latest contribution to this task, integrating neural network language models in the decoding process of three state-of-the-art systems: one based on bidirectional recurrent neural networks, another based on hybrid hidden Markov models and, finally, a combination of both. Experimental results obtained on the IAM off-line database demonstrate that consistent word error rate reductions can be achieved with neural network language models when compared with statistical N-gram language models on the three tested systems. The best word error rate, 16.1%, reported with ROVER combination of systems using neural network language models significantly outperforms current benchmark results for the IAM database., The authors wish to acknowledge the anonymous reviewers for their detailed and helpful comments to the paper. We also thank Alex Graves for kindly providing us with the BLSTM Neural Network source code. This work has been supported by the European project FP7-PEOPLE-2008-IAPP: 230653, the Spanish Government under project TIN2010-18958, as well as by the Swiss National Science Foundation (Project CRSI22_125220).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Handwriting recognition in historical documents using very large vocabularies

  • Frinken, Volkmar
  • Fischer, Andreas
  • Martínez-Hinarejos, Carlos-D.|||0000-0002-6139-2891
Language models are used in automatic transcription system to resolve ambiguities. This is done by limiting the vocabulary of words that can be recognized as well as estimating the n-gram probability of the words in the given text. In the context of historical documents, a non-unified spelling and the limited amount of written text pose a substantial problem for the selection of the recognizable vocabulary as well as the computation of the word probabilities. In this paper we propose for the transcription of historical Spanish text to keep the corpus for the n-gram limited to a sample of the target text, but expand the vocabulary with words gathered from external resources. We analyze the performance of such a transcription system with different sizes of external vocabularies and demonstrate the applicability and the significant increase in recognition accuracy of using up to 300 thousand external words., This work has been supported by the European project FP7-PEOPLE-2008-IAPP: 230653 the European Research Council’s Advanced Grant ERC-2010-AdG 20100407, the Spanish R&D projects TIN2009-14633-C03-03, RYC-2009-05031, TIN2011-24631, TIN2012-37475-C02-02, MITTRAL (TIN2009-14633-C03-01), Active2Trans (TIN2012-31723) as well as the Swiss National Science Foundation fellowship project PBBEP2_141453.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The ESPOSALLES database: An ancient marriage license corpus for off-line handwriting recognition

  • Romero Gómez, Verónica
  • Fornés, Alicia
  • Serrano Martínez-Santos, Nicolás
  • Frinken, Volkmar
  • Lladós, Josep
  • Sánchez Peiró, Joan Andreu|||0000-0003-0423-2020
  • Toselli, Alejandro Héctor|||0000-0001-6955-9249
  • Vidal, Enrique|||0000-0003-4579-5196
[EN] Historical records of daily activities provide intriguing insights into the life of our ancestors, useful for demography studies and genealogical research. Automatic processing of historical documents, however, has mostly been focused on single works of literature and less on social records, which tend to have a distinct layout, structure, and vocabulary. Such information is usually collected by expert demographers that devote a lot of time to manually transcribe them. This paper presents a new database, compiled from a marriage license books collection, to support research in automatic handwriting recognition for historical documents containing social records. Marriage license books are documents that were used for centuries by ecclesiastical institutions to register marriage licenses. Books from this collection are handwritten and span nearly half a millennium until the beginning of the 20th century. In addition, a study is presented about the capability of state-of-the-art handwritten text recognition systems, when applied to the presented database. Baseline results are reported for reference in future studies. © 2012 Elsevier Ltd. All rights reserved., Work supported by the EC (FEDER/FSE) and the Spanish MEC/MICINN under the MIPRCV ‘‘Consolider Ingenio 2010’’ program (CSD2007-00018), MITTRAL (TIN2009-14633-C03-01) and KEDIHC ((TIN2009-14633-C03-03) projects. This work has been partially supported by the European Research Council Advanced Grant (ERC-2010-AdG-20100407: 269796-5CofM) and the European seventh framework project (FP7-PEOPLE-2008-IAPP: 230653-ADAO). Also supported by the Generalitat Valenciana under grant Prometeo/2009/014 and FPU AP2007-02867, and by the Universitat Politecnica de Val encia (PAID-05-11). We would also like to thank the Center for Demographic Studies (UAB) and the Cathedral of Barcelona.