Introducción a IMPACT. Isabel Bordes Cabrera

• 1. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.INTRODUCCIN A IMPACTSESIN DE DEMOSTRACIN IMPACTBiblioteca Nacional de Espaa, 5 de octubre de 2011

2. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.AntecedentesEl texto no digitalizado es virtualmente invisibleDisponibilidad de material digitalizado demasiado lento, pequeas cantidades ypocas fuentesTecnologa OCR (reconocimiento ptico de caracteres) no resultados satisfactoriospara documentos histricosFalta de conocimiento institucional y especializacinineficiencia y reinventar larueda 2 3. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Texto histrico: tpicos resultados OCRVVt Venetien den 1.Junij, Anno 1618.DJgn i f paffato te S aJifeert mo?en/bah .)etgiuotbciraetail)i.r/JtmelchontDecht te /sbnbe bele btr felbrr geiufttceert baer bnber eeniglje jprant o^fen/bie ftcb .metbeSpaenfcbeu enbeeemgljen bifet Cbeiiupcen berbonbru befe 3 4. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Retos OCRTintas traspasadasManchasEncuadernacionesOrificios/desgarrosErrores de impresinSellosPlieguesSuciedadAnimalesEntintado dbilReparacionesCalidad del papelLaminacin4 5. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Retos lingsticos: variantes ortogrficas, variantes degrafa, formas conjugadas, declinadasy ms Variantes histricas para la forma verbal hubo: hubo, huvo, uvo, vuo, uuo, huuo Variantes histricas para la palabra holandesa wereld (mundo): werelt weerelt wereld weerelds wereldt werelden weereld werrelts waerelds weerlyt wereldts vveerelts waereld weerelden waerelden weerlt werlt werelds sweerels zwerlys swarels swerelts werelts swerrels weirelts tsweerelds werret vverelt werlts werrelt worreld werlden wareld weirelt weireld waerelt werreld werld vvereld weerelts werlde tswerels werreldts weereldt wereldje waereldje weurlt wald weled 5 6. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Reto institucional: falta de conocimiento y experiencia ineficiencia6 7. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Cmo afrontar estos retos: La visin de IMPACT Digitalizar los textos impresos histricosen Europa mejor, ms rpido , msbarato Proporcionar herramientas, servicios einstalaciones para continuar avanzandoen este campo 7 8. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.El proyecto en cifrasApoyado por la Comunidad Europea a travs de la lnea ICT del FP7Coordinacin Biblioteca Nacional de Pases Bajos (KB)Tipo de proyecto: Proyecto de integracin a Gran EscalaFinanciacin de la UE : 11 500 000Fecha de inicio: 1 enero 2008Duracin: 48 mesesA partir de 2012: Centro de competencia sostenibleContacto: [email protected] web: www.impact-project.eu 8 9. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Consorcio IMPACT - InicialBibliotecas Universidades y Centros de Investigacin National Library of the Netherlands (KB) Dutch Institute for Lexicology (INL) The British Library (BL) National Centre for Scientific Research Bibliothque nationale de France (BNF) Demokritos (NCSR) German National Library (DNB) University of Salford (USAL) Bavarian State Library (BSB) University of Munich (CIS group) Gttingen State and University Library University of Innsbruck (InfMath group) (UGOE) University of Bath (UKOLN) Austrian National Library (ONB) University of Innsbruck Library (UIBK) Proveedores comerciales IBM (Haifa Research Lab) ABBYY (Moscow) 9 10. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.OBJETIVOSMejora significativa de la digitalizacin masiva del texto impreso histricos mediante:Innovacin del software OCR y tecnologas en torno a los idiomasCompartir conocimientos y poner en pie habilidades a nivel europeoGarantizar existencia a largo plazo herramientas y servicios al finalizar el proyectoConseguir el intercambio de procesos o datos entre sistemas heterogneos(interoperabilidad)Proporcionar una Gua de Mejores Prcticas en el mbito de la digitalizacinAtender las necesidades de usuarios en diversos idiomas 10 11. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. Inicio 2010 nuevos socios i.a.: ... 23 Biblioteca Nacional de EspaaBNE ES24 Fundacin Biblioteca Virtual Miguel de Cervantes BVC ES25 Pozna Supercomputing and Networking Center PSNC PL BNE/UA = marzo de 2010 11 12. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 12 13. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Extensin de objetivosUtilizar las herramientas de IMPACT para construir un lxico eficiente para idiomasfuera de los objetivos iniciales de IMPACT Inicialmente: ingls, alemn, holands Dp: lenguas romance y eslavas ej espaolDemostrar y difundir los resultados del proyecto en Europa del Sur y del Este, yapoyar la adquisicin de conocimientos en digitalizacin en dichos pasesReforzar la cooperacin y mejorar la explotacin de las sinergias de I+D sobre ICTa lo largo de la Unin EuropeaEstablecer vnculos estratgicos con el fin de avanzar en:el acceso al conocimiento,desarrollo de estndares ysoluciones interoperables 13 14. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Estructura del proyecto: 4 subproyectosOPERATIONAL CONTEXT (OC)Requisitos, documentacin wfs y medicionesMejores prcticas y directricesMarco de trabajo e integracin tcnica - interoperabilidad TEXT RECOGNITION (TR) -5 WG ENHANCEMENT & ENRICHMENT (EE) Pre-procesamiento y segmentacinCorreccin colaborativaOCR adaptativo y experimental Lexica e ndices geogrficos Modelos y diccionariosMetadatos estructurales CAPACITY BUILDING (CB)Publicacin recursos Formacin y apoyo Demostraciones 14 15. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Herramientas de OCRTecnologas para la extraccin de texto en formato digital a partir de una pginaMotor de OCR adaptativo: Clave del proyecto IMPACT, software de primera lneaadaptado especficamente a las necesidades de las bibliotecas y que se adapta durante elproceso de OCR, integrando otras herramientas: Image enhancement toolkit Segmentation toolkit Mdulos de post-correccin Otros motores OCRPrototipos y herramientas experimentales Prototipo OCR sobre textos escritos a mquina (Typewritten OCR prototype) Wordspotting OC TREECB 15 16. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Herramientas de enriquecimiento (tecnologalingstica)Conseguir que los resultados de OCR sean ms precisos y accesibles Correccin colaborativa Sistema de correccin colaborativa a travs de web: plataforma web, adecuada para participacin voluntaria masiva, valida y corrige resultados OCR. La primera herramienta de este tipo que se relaciona directamente a un motor de OCRLexicon e ndices geogrficosLxicos generales y NER (nnll-uk-ger)Herramientas para proporcionar medios que permitan superar barrera lingsticahistricaMetadatos estructuralesFEP: un conjunto de servicios web que pueden explotarse para detectar y etiquetarautomticamente metadatos estructurales del material escaneadoOC TR EECB 16 17. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Herramientas y servicios estratgicosSitio web proporciona acceso a todos los resultados del proyecto y constituye el ncleo de una red virtual formadapor todos los centros de competencia europeos en materia de digitalizacin y las actividades de investigacinasociadasUn conjunto de herramientas de Apoyo a la Toma de Decisiones que puede utilizarse para inciar,organizar, gestionar y evaluar el coste de los proyectos de digitalizacin masivaRecursos de aprendizaje (learning resource toolbox) con directrices de funcionamiento que sirvande gua para la implementacin real de todas las herramientas producidas en el proyecto.Formacin y apoyo Sistema Help Desk peticiones, ayuda entre usuarios, socios del proyecto, centros de competencia de digitalizacin. Programa de formacin sobre temas y tecnologas en torno a la digitalizacin a gran escala con una amplia gama de documentacin de apoyo disponible a travs del sitio web del proyectoDemosOC TR EE CB 17 18. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Biblioteca Nacional de Espaa (CB)Entr a formar parte en marzo de 2010, II Fase del proyecto , en colaboracin conla Universidad de AlicanteAportacin de ms de 60.000 imgenes (UA-BVMC) y subida a PRIMA (md):Escala grises, color, 300 dpi, TIFFObras anteriores XIX ej: Teatro Siglo de OroObras cuyo OCR no est ofreciendo buenos resultados En 3 sets: Development (Lexicon Building) Evaluation - DemonstrationProbar las funcionalidades realizadas por los desarrolladores participantesDemonstration datasetLotes, repositorio resultadosDifundir el proyecto (presentaciones a nivel interno, FESABID, sesin externa) 18 19. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Universidad de Alicante (EE-CB)Colaboracin en carga en el servidor de imgenes PRIMAGestin de contrato GroundTruth (evaluacin), control de calidad y subida a PRIMADifundir el proyectoCreacin de lxico espaol antiguo para conectar a las soluciones de OCR 19 20. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. Herramientas del entorno de trabajo 20 21. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.PRIMARepositorio de imgenes aportadas por las bibliotecas participantes que sirven de ejemplopara los diversos planes proyectadosHome 21 22. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.PRIMA: Groundtruth-formato PAGE22 23. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.MyExperimentPlataforma o entorno de desarrollo donde investigadores intercambian ideas, desarrollos yplanesHome 23 24. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 24 25. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.TAVERNAHerramienta creada por un Grupo de Trabajo (myGrid) , de cdigo libre (JAVA), que permite eldiseo y ejecucin de Flujos de Trabajos Servicio Web: Interoperabilidad Facilidad acceso Dinamismo Flujos simples & complejos25 26. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.TAVERNA: flujos simples o complejos 26 27. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Funcionamiento Taverna en Impact 27 28. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 28 29. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 29 30. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 30 31. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 31 32. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 32 33. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 33 34. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 34 35. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 35 36. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 36 37. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 37 38. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 38 39. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. 39 40. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Cmo estar al da sobre IMPACTEn TwitterUnin al LinkedIn groupVisita el blog y el websiteLista de correo del proyecto para recibir el boletn trimestralCanal YoutubePara cualquier pregunta feedback: contactar con IMPACT helpdesk 40 41. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Twitter: @impactocr, #impactprojectTweets sobre ltimasnoticias en IMPACT(Re)tweets de noticiasen torno a ladigitalizacin,tecnologa OCR ylingsticaTweets en vivodurante conferencias,talleres 41 42. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.LinkedIn group: IMPACT Improving Access to Text Comunidad online Canal para obtener feedback sobre los entregables IMPACT (versiones posteriores) Discusiones sobre digitalizacin, tecnologa OCR y lingstica42 43. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.IMPACT blog: impactocr.wordpress.com Utilizado para talleres y das de demostracin Incorporacin dePresentacionesVdeos43 44. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.www.impact-project.eu 44 45. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Boletn trimestral IMPACTSecciones: Novedades Especial en torno a una herramienta Fotos e informes sobre contribuciones de IMPACT a conferencias Entrevistas equipo humano IMPACT 45 46. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.Canal de IMPACT en Youtubewww.youtube.com/user/theimpactproject Breves introducciones sobre cada subproyecto y cada uno de los lderes de los distintos work package Vdeo de presentacin de IMPACT elaborado por la KB Video de IBM sobre CONCERT46 47. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands. IMPACT Helpdeskwww.impact-project.eu/helpdesk47 48. IMPACT is supported by the European Community under the FP7 ICT Work Programme. The project is coordinated by the National Library of the Netherlands.GRACIAS POR SU ATENCINIsabel Bordes CabreraJefe de Servicio de Biblioteca Digital, [email protected] 48