Skip to main content

HITITA: HERRAMIENTA INTERACTIVA PARA LA TRANSCRIPCION DE IMAGENES DE TEXTOS ANTIGUOS

TIN2010-18958

•
Funding agency name Ministerio de Ciencia e Innovación
Funding agency acronym MICINN
Program Programa Nacional de Proyectos de Investigación Fundamental
Subprogram Subprograma de Proyectos de Investigación Fundamental no Orientada
Call Proyectos I+D
Call year 2010
State Plan 2008-2011 Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica
Management unit Subdirección General de Proyectos de Investigación
Beneficiary centre UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication 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
Publication Contribución de congreso (ConferenceOutput)

esCam: A Mobile Application to Capture and Enhance Text Images

  • Pastor Pellicer, Joan
  • Adelantado Torres, José Luís
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
[EN] Taking high resolution photos with mobile devices anytime anywhere is becoming increasingly common. Therefore, images of all kinds of text documents are recorded. This work presents esCam, an application for Android platform, whose goal is to preprocess the images of those text documents, in particular, perspective correction and image cleaning and enhancing. What truly differentiates our application is that esCam focuses on treatment of text that may appear in the image, using neural networks. These preprocessing steps are needed to make easier the digitalization and also to benefit subsequent steps such as document analysis and text recognition., This work has been partially supported by the Spanish Government project TIN2010-18958.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Human or computer assisted interactive transcription: automated text recognition, text annotation, and scholarly edition in the twenty-first century, Transcripción humana o asistencia a la transcripción automática interactiva: reconocimiento automático del texto, anotación y edición erudita en el siglo XXI

  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Vilar Torres, Juan Miguel
  • Llorens, D.
  • Marzal Varó, Andrés
  • Prat, F.
  • Zamora Martínez, Francisco Julián
[EN] Computer assisted transcription tools can speed up the initial process of reading and transcribing texts. At the same time, new annotation tools open new ways of accessing the text in its graphical form. The balance and value of each method still needs to be explored. STATE, a complete assisted transcription system for ancient documents, was presented to the audience of the 2013 International Medieval Congress at Leeds. The system offers a multimodal interaction environment to assist humans in transcribing ancient documents: the user can type, write on the screen with a stylus, or utter a word. When one of these actions is used to correct an erroneous word, the system uses this new information to look for other mistakes in the rest of the line. The system is modular, composed of different parts: one part creates projects from a set of images of documents, another part controls an automatic transcription system, and the third part allows the user to interact with the transcriptions and easily correct them as needed. This division of labour allows great flexibility for organising the work in a team of transcribers., [ES] Las herramientas de ayuda a la transcripción automática pueden acelerar el proceso inicial de la lectura y transcripción de textos. Al mismo tiempo, las nuevas herramientas de anotación aportan nuevas formas de acceder al texto en su forma original gráfica. Sin embargo, todavía es necesario evaluar las bondades y capacidades de los distintos métodos. STATE, un completo sistema de asistencia a la transcripción de documentos antiguos, se presentó a la audiencia del International Medieval Congress de 2013 celebrado en Leeds. El sistema ofrece un entorno de interacción multimodal para ayudar a las personas en la transcripción de documentos antiguos: el usuario puede teclear, escribir en la pantalla con un lápiz óptico o corregir usando la voz. Cada vez que el usuario cambia de esta forma una palabra, el sistema utiliza la corrección para buscar errores en el resto de la línea. El sistema está dividido en diferentes módulos: uno crea proyectos a partir de un conjunto de imágenes de documentos, otro módulo controla el sistema de transcripción automática, y un tercer módulo permite al usuario interactuar con las transcripciones y corregirlas fácilmente cuando sea necesario. Esta división de las tareas permite una gran flexibilidad para organizar el trabajo de los transcriptores en equipo., Work supported by the Spanish Government (TIN2010-18958) and the Generalitat Valenciana (Prometeo/2010/028)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Handwriting Normalization by Zone Estimation using HMM/ANNs

  • Pastor Pellicer, Joan
  • Zamora Martínez, Francisco
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
Offline handwritten text recognition requires several preprocessing stages. Many different preprocessing techniques have been proposed in the literature based either on geometrical heuristics or on statistical models. Unfortunately, these approaches usually fail when dealing with short sentences or isolated words. One statistical technique for text line preprocessing is based on the detection and classification of local extrema points, by means of neural networks, to determine the reference lines delimiting the different zones. This technique depends on a sufficient amount of local extrema and, relating its robustness, a single bad classified extrema point may lead to undesirable results. This paper proposes a novel method to normalize handwritten text lines based on a supervised statistical model which takes into account all pixels instead of just the local extrema. A Hidden Markov Model hybridized with an Artificial Neural Network is applied column-wise in order to segment each column of the handwritten line into three zones. The reference lines obtained in this way are used to normalize the image afterwards. The technique has been empirically tested on the IAM offline database., This work has been supported by the Spanish Government under project TIN2010-18958
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Capítulo o parte de un libro (BookPart)

F-Measure as the error function to train neural networks

  • Pastor Pellicer, Joan
  • Zamora Martínez, Francisco Julián
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
Imbalance datasets impose serious problems in machine learning. For many tasks characterized by imbalanced data, the F-Measure seems more appropiate than the Mean Square Error or other error measures. This paper studies the use of F-Measure as the training criterion for Neural Networks by integrating it in the Error-Backpropagation algorithm. This novel training criterion has been validated empirically on a real task for which F-Measure is typically applied to evaluate the quality. The task consists in cleaning and enhancing ancient document images which is performed, in this work, by means of neural filters., This work has been partially supported by MICINN project HITITA (TIN2010-18958) and by the FPI-MICINN (BES-2011-046167) scholarship from Ministerio de Ciencia e Innovación, Gobierno de España.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Complete Text Line Extraction with Convolutional Neural Networks and Watershed Transform

  • Pastor Pellicer, Joan
  • Zeshan Afzal, Muhammad
  • Liwicki, Marcus
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
We present a novel Convolutional Neural Network- based method for the extraction of text lines, which consists of an initial Layout Analysis followed by the estimation of the Main Body Area (i.e., the text area between the baseline and the corpus line) for each text line. Finally, a region-based method using watershed transform is performed on the map of the Main Body Area for extracting the resulting lines. We have evaluated the new system on the IAM-HisDB, a publicly available dataset containing historical documents, outperforming existing learning-based text line extraction methods, which consider the problem as pixel labelling problem into text and non-text regions., This work has been partially supported by the the Spanish Government project TIN2010-18958.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

A combined Convolutional Neural Network and Dynamic Programming approach for text line normalization

  • Pastor Pellicer, Joan
  • Zamora Martínez, Francisco
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
[EN] This work proposes a new normalization algorithm for handwritten text lines based on the use of Convolutional Neural Networks trained to classify pixels of the scanned text line as belonging to the main body area. The reference lines of the text line are obtained from these local estimates by means of Dynamic Programming. The obtained reference lines are used to normalize the text line images. Experimental results on the IAM offline database demonstrates the feasibility of this approach., This work has been supported by the Spanish Government under project TIN2010-18958.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Speech Interaction in a Multimodal Tool for Handwritten Text Transcription

  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
  • España Boquera, Salvador|||0000-0002-0335-9997
  • Llorens, David
  • Marzal Varó, Andrés
  • Prat, Federico
  • Vilar Torres, Juan Miguel
  • Zamora Martínez, Francisco
[EN] STATE is a multimodal tool for document processing and text transcription. Its graphical front-end can be easily connected to different text recognition back-ends. New features and improvements are presented in this work: the interactive correction of one word in the transcribed line has been improved to reestimate the entire transcription line using the user feedback and speech input has been integrated in the multimodal interface enabling the user to also utter the word to be corrected, giving the user the possibility to use the interface according to her preferences or the task at hand. Thus, at the current version of STATE, the user can type, write on the screen with a stylus, or utter the incorrectly recognized word, and then, the system uses the user feedback in any of the proposed modalities to reestimate the transcribed line so as to hopefully correct other errors which could be caused by the mistaken word the user has corrected., Work partially supported by the Spanish Government (TIN2010-18958 and Consolider Ingenio 2010 CSD2007-00018), the Generalitat Valenciana (Prometeo/2010/028), and Fundació Caixa CastellóBancaixa (P11B2009-48)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Combining Learned Script Points and Combinatorial Optimization for Text Line Extraction

  • Pastor Pellicer, Joan
  • Garz, Angelika
  • Ingold, Rolf
  • Castro-Bleda, Maria Jose|||0000-0003-1001-8258
Complex layouts, curved text lines, heterogeneous background, noise, and clutter still render text line extraction in the context of historical documents a challenging task where traditional methods do not excel. We propose a novel text line extraction method with two contributions: first, text-specific interest points extracted by supervised machine learning; and second, reformulating the problem of bottom-up text line aggregation as noise-robust combinatorial optimization. In a final step, unsupervised clustering eliminates invalid text lines. Building the method on top of interest points and posing aggregation as global optimization problem, we can detect text lines with arbitrary orientation and curvature, and are robust to noise and clutter. Experimental evaluations on the IAM Saint Gall dataset show promising results., This work has been supported by the Swiss National Science Foundation project 205120 150173 and the Spanish Government project TIN2010-18958.