- -

HMM word graph based keyword spotting in handwritten document images

RiuNet: Institutional repository of the Polithecnic University of Valencia

Share/Send to

Cited by

Statistics

HMM word graph based keyword spotting in handwritten document images

Show simple item record

Files in this item

dc.contributor.author Toselli, Alejandro Héctor es_ES
dc.contributor.author Vidal, Enrique es_ES
dc.contributor.author Romero, Verónica es_ES
dc.contributor.author Frinken, Volkmar es_ES
dc.date.accessioned 2019-05-24T20:02:40Z
dc.date.available 2019-05-24T20:02:40Z
dc.date.issued 2016 es_ES
dc.identifier.issn 0020-0255 es_ES
dc.identifier.uri http://hdl.handle.net/10251/121061
dc.description.abstract [EN] Line-level keyword spotting (KWS) is presented on the basis of frame-level word posterior probabilities. These posteriors are obtained using word graphs derived from the recogni- tion process of a full-fledged handwritten text recognizer based on hidden Markov models and N-gram language models. This approach has several advantages. First, since it uses a holistic, segmentation-free technology, it does not require any kind of word or charac- ter segmentation. Second, the use of language models allows the context of each spotted word to be taken into account, thereby considerably increasing KWS accuracy. And third, the proposed KWS scores are based on true posterior probabilities, taking into account all (or most) possible word segmentations of the input image. These scores are properly bounded and normalized. This mathematically clean formulation lends itself to smooth, threshold-based keyword queries which, in turn, permit comfortable trade-offs between search precision and recall. Experiments are carried out on several historic collections of handwritten text images, as well as a well-known data set of modern English handwrit- ten text. According to the empirical results, the proposed approach achieves KWS results comparable to those obtained with the recently-introduced "BLSTM neural networks KWS" approach and clearly outperform the popular, state-of-the-art "Filler HMM" KWS method. Overall, the results clearly support all the above-claimed advantages of the proposed ap- proach. es_ES
dc.description.sponsorship This work has been partially supported by the Generalitat Valenciana under the Prometeo/2009/014 project grant ALMA-MATER, and through the EU projects: HIMANIS (JPICH programme, Spanish grant Ref. PCIN-2015-068) and READ (Horizon 2020 programme, grant Ref. 674943). es_ES
dc.language Inglés es_ES
dc.publisher Elsevier es_ES
dc.relation GV/PROMETEO/2009/014 es_ES
dc.relation MINECO/PCIN-2015-068 es_ES
dc.relation.ispartof Information Sciences es_ES
dc.rights Reconocimiento - No comercial - Sin obra derivada (by-nc-nd) es_ES
dc.subject Keyword spotting es_ES
dc.subject Handwritten text recognition es_ES
dc.subject Word graph es_ES
dc.subject Posterior probability es_ES
dc.subject Confidence score es_ES
dc.subject.classification ESTADISTICA E INVESTIGACION OPERATIVA es_ES
dc.subject.classification LENGUAJES Y SISTEMAS INFORMATICOS es_ES
dc.title HMM word graph based keyword spotting in handwritten document images es_ES
dc.type Artículo es_ES
dc.identifier.doi 10.1016/j.ins.2016.07.063 es_ES
dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/674943/EU es_ES
dc.rights.accessRights Abierto es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Sistemas Informáticos y Computación - Departament de Sistemes Informàtics i Computació es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Estadística e Investigación Operativa Aplicadas y Calidad - Departament d'Estadística i Investigació Operativa Aplicades i Qualitat es_ES
dc.description.bibliographicCitation Toselli, AH.; Vidal, E.; Romero, V.; Frinken, V. (2016). HMM word graph based keyword spotting in handwritten document images. Information Sciences. 370:497-518. https://doi.org/10.1016/j.ins.2016.07.063 es_ES
dc.description.accrualMethod S es_ES
dc.relation.publisherversion http://dx.doi.org/10.1016/j.ins.2016.07.063 es_ES
dc.description.upvformatpinicio 497 es_ES
dc.description.upvformatpfin 518 es_ES
dc.type.version info:eu-repo/semantics/publishedVersion es_ES
dc.description.volume 370 es_ES
dc.relation.pasarela S\338507 es_ES
dc.contributor.funder Generalitat Valenciana es_ES
dc.contributor.funder Ministerio de Economía y Empresa es_ES
dc.contributor.funder European Commission es_ES


This item appears in the following Collection(s)

Show simple item record