scispace - formally typeset
Search or ask a question

Showing papers on "Handwriting recognition published in 2007"


Journal ArticleDOI
TL;DR: New and very effective techniques for automatic writer identification and verification that use probability distribution functions (PDFs) extracted from the handwriting images to characterize writer individuality are developed.
Abstract: The identification of a person on the basis of scanned images of handwriting is a useful biometric modality with application in forensic and historic document analysis and constitutes an exemplary study area within the research field of behavioral biometrics. We developed new and very effective techniques for automatic writer identification and verification that use probability distribution functions (PDFs) extracted from the handwriting images to characterize writer individuality. A defining property of our methods is that they are designed to be independent of the textual content of the handwritten samples. Our methods operate at two levels of analysis: the texture level and the character-shape (allograph) level. At the texture level, we use contour-based joint directional PDFs that encode orientation and curvature information to give an intimate characterization of individual handwriting style. In our analysis at the allograph level, the writer is considered to be characterized by a stochastic pattern generator of ink-trace fragments, or graphemes. The PDF of these simple shapes in a given handwriting sample is characteristic for the writer and is computed using a common shape codebook obtained by grapheme clustering. Combining multiple features (directional, grapheme, and run-length PDFs) yields increased writer identification and verification performance. The proposed methods are applicable to free-style handwriting (both cursive and isolated) and have practical feasibility, under the assumption that a few text lines of handwritten material are available in order to obtain reliable probability estimates

468 citations


Journal ArticleDOI
TL;DR: It is shown in a subset of the George Washington collection that such a word spotting technique can outperform a Hidden Markov Model word-based recognition technique in terms of word error rates.
Abstract: Searching and indexing historical handwritten collections are a very challenging problem. We describe an approach called word spotting which involves grouping word images into clusters of similar words by using image matching to find similarity. By annotating “interesting” clusters, an index that links words to the locations where they occur can be built automatically. Image similarities computed using a number of different techniques including dynamic time warping are compared. The word similarities are then used for clustering using both K-means and agglomerative clustering techniques. It is shown in a subset of the George Washington collection that such a word spotting technique can outperform a Hidden Markov Model word-based recognition technique in terms of word error rates.

368 citations


Proceedings Article
03 Dec 2007
TL;DR: A system capable of directly transcribing raw online handwriting data is described, consisting of an advanced recurrent neural network with an output layer designed for sequence labelling, combined with a probabilistic language model.
Abstract: In online handwriting recognition the trajectory of the pen is recorded during writing. Although the trajectory provides a compact and complete representation of the written output, it is hard to transcribe directly, because each letter is spread over many pen locations. Most recognition systems therefore employ sophisticated preprocessing techniques to put the inputs into a more localised form. However these techniques require considerable human effort, and are specific to particular languages and alphabets. This paper describes a system capable of directly transcribing raw online handwriting data. The system consists of an advanced recurrent neural network with an output layer designed for sequence labelling, combined with a probabilistic language model. In experiments on an unconstrained online database, we record excellent results using either raw or preprocessed data, well outperforming a state-of-the-art HMM based system in both cases.

262 citations


Journal ArticleDOI
TL;DR: It is shown experimentally that the proposed nonlinear image deformation models performs very well for four different handwritten digit recognition tasks and for the classification of medical images, thus showing high generalization capacity.
Abstract: We present the application of different nonlinear image deformation models to the task of image recognition The deformation models are especially suited for local changes as they often occur in the presence of image object variability We show that, among the discussed models, there is one approach that combines simplicity of implementation, low-computational complexity, and highly competitive performance across various real-world image recognition tasks We show experimentally that the model performs very well for four different handwritten digit recognition tasks and for the classification of medical images, thus showing high generalization capacity In particular, an error rate of 054 percent on the MNIST benchmark is achieved, as well as the lowest reported error rate, specifically 126 percent, in the 2005 international ImageCLEF evaluation of medical image specifically categorization

257 citations


Posted Content
TL;DR: In this paper, multi-dimensional recurrent neural networks (MDRNNs) have been proposed for image segmentation, which can be used for vision, video processing, medical imaging and many other areas.
Abstract: Recurrent neural networks (RNNs) have proved effective at one dimensional sequence learning tasks, such as speech and online handwriting recognition. Some of the properties that make RNNs suitable for such tasks, for example robustness to input warping, and the ability to access contextual information, are also desirable in multidimensional domains. However, there has so far been no direct way of applying RNNs to data with more than one spatio-temporal dimension. This paper introduces multi-dimensional recurrent neural networks (MDRNNs), thereby extending the potential applicability of RNNs to vision, video processing, medical imaging and many other areas, while avoiding the scaling problems that have plagued other multi-dimensional models. Experimental results are provided for two image segmentation tasks.

220 citations


Book ChapterDOI
09 Sep 2007
TL;DR: Multi-dimensional recurrent neural networks are introduced, thereby extending the potential applicability of RNNs to vision, video processing, medical imaging and many other areas, while avoiding the scaling problems that have plagued other multi-dimensional models.
Abstract: Recurrent neural networks (RNNs) have proved effective at one dimensional sequence learning tasks, such as speech and online handwriting recognition. Some of the properties that make RNNs suitable for such tasks, for example robustness to input warping, and the ability to access contextual information, are also desirable in multi-dimensional domains. However, there has so far been no direct way of applying RNNs to data with more than one spatio-temporal dimension. This paper introduces multi-dimensional recurrent neural networks, thereby extending the potential applicability of RNNs to vision, video processing, medical imaging and many other areas, while avoiding the scaling problems that have plagued other multi-dimensional models. Experimental results are provided for two image segmentation tasks.

215 citations


Proceedings Article
01 Jan 2007
TL;DR: A new connectionist approach to on-line handwriting recognition and address in particular the problem of recognizing handwritten whiteboard notes using a recently introduced objective function, known as Connectionist Temporal Classification (CTC), that directly trains the network to label unsegmented sequence data.
Abstract: In this paper we introduce a new connectionist approach to on-line handwriting recognition and address in particular the problem of recognizing handwritten whiteboard notes. The approach uses a bidirectional recurrent neural network with the long short-term memory architecture. We use a recently introduced objective function, known as Connectionist Temporal Classification (CTC), that directly trains the network to label unsegmented sequence data. Our new system achieves a word recognition rate of 74.0%, compared with 65.4% using a previously developed HMMbased recognition system.

204 citations


Proceedings ArticleDOI
23 Sep 2007
TL;DR: The aim of this contest was to use well established evaluation practices and procedures in order to record recent advances in off-line handwriting segmentation and to test and compare all submitted algorithms for handwritten document segmentation in realistic circumstances.
Abstract: This paper presents the results of the handwriting segmentation contest that was organized in the context of ICDAR2007. The aim of this contest was to use well established evaluation practices and procedures in order to record recent advances in off-line handwriting segmentation. Two benchmarking datasets (one for text line and one for word segmentation) were used in a common evaluation platform in order to test and compare all submitted algorithms for handwritten document segmentation in realistic circumstances. The results of the evaluation of five algorithms submitted by participants as well as of two state-of-the-art algorithms are presented. The performance evaluation method is based on counting the number of matches between the text lines or words detected by the algorithms and the text line or words of the ground truth.

170 citations


Journal ArticleDOI
TL;DR: The statistics show that the HIT-MW database has an excellent representation of the real handwriting and many new applications concerning real handwriting recognition can be supported by the database.
Abstract: A Chinese handwriting database named HIT-MW is presented to facilitate the offline Chinese handwritten text recognition. Both the writers and the texts for handcopying are carefully sampled with a systematic scheme. To collect naturally written handwriting, forms are distributed by postal mail or middleman instead of face to face. The current version of HIT-MW includes 853 forms and 186,444 characters that are produced under an unconstrained condition without preprinted character boxes. The statistics show that the database has an excellent representation of the real handwriting. Many new applications concerning real handwriting recognition can be supported by the database.

142 citations


Journal ArticleDOI
TL;DR: A new feature extraction approach, called normalization-cooperated gradient feature (NCGF) extraction, which maps the gradient direction elements of original image to direction planes without generating the normalized image and can be combined with various normalization methods.
Abstract: The gradient direction histogram feature has shown superior performance in character recognition. To alleviate the effect of stroke direction distortion caused by shape normalization and provide higher recognition accuracies, we propose a new feature extraction approach, called normalization-cooperated gradient feature (NCGF) extraction, which maps the gradient direction elements of original image to direction planes without generating the normalized image and can be combined with various normalization methods. Experiments on handwritten Japanese and Chinese character databases show that, compared to normalization-based gradient feature, the NCGF reduces the recognition error rate by factors ranging from 8.63 percent to 14.97 percent with high confidence of significance when combined with pseudo-two-dimensional normalization.

141 citations


Proceedings ArticleDOI
23 Sep 2007
TL;DR: The results show that the methods developed and tested on Western script in recent years are very effective and the conclusions drawn in previous studies remain valid also on Arabic script.
Abstract: In this paper, we evaluate the performance on Arabic handwriting of the text-independent writer identification methods that we developed and tested on Western script in recent years. We use the IFN/ENIT data in the experiments reported here and our tests involve 350 writers. The results show that our methods are very effective and the conclusions drawn in previous studies remain valid also on Arabic script. High performance is achieved by combining textural features (joint directional probability distributions) with allographic features (grapheme-emission distributions).

Proceedings ArticleDOI
23 Sep 2007
TL;DR: An online handwritten symbol recognition system for Telugu, a widely spoken language in India, is presented based on hidden Markov models (HMM) and uses a combination of time-domain and frequency-domain features.
Abstract: In this paper we present an online handwritten symbol recognition system for Telugu, a widely spoken language in India. The system is based on hidden Markov models (HMM) and uses a combination of time-domain and frequency-domain features. The system gives top-1 accuracy of 91.6% and top-5 accuracy of 98.7% on a dataset containing 29,158 train samples and 9,235 test samples. We also introduce a cost-effective and natural data collection procedure based on ACECADreg Digimemoreg and describe its usage in building a Telugu handwriting dataset.

Journal ArticleDOI
TL;DR: An off-line, text independent system for writer identification and verification of handwritten text lines using Hidden Markov Model (HMM) based recognizers is presented.
Abstract: In this paper, an off-line, text independent system for writer identification and verification of handwritten text lines using Hidden Markov Model (HMM) based recognizers is presented. For each writer, an individual recognizer is built and trained on text lines of that writer. This results in a number of recognizers, each of which is an expert on the handwriting of exactly one writer. In the identification and verification phase, a text line of unknown origin is presented to each of these recognizers and each one returns a transcription that includes the log-likelihood score for the generated output. These scores are sorted and the resulting ranking is used for both identification and verification. Several confidence measures are defined on this ranking. The proposed writer identification and verification system is evaluated using different experimental setups.

Proceedings ArticleDOI
03 Dec 2007
TL;DR: In this article, a modified exponential membership function fitted to the fuzzy sets derived from features consisting of normalized distances obtained using the Box approach was used for Hindi character recognition. But, the accuracy of the overall recognition rate was only 90.65%.
Abstract: This paper presents the recognition of handwritten Hindi Characters based on the modified exponential membership function fitted to the fuzzy sets derived from features consisting of normalized distances obtained using the Box approach. The exponential membership function is modified by two structural parameters that are estimated by optimizing an objective function that includes the entropy and error function. A Reuse Policy that provides guidance from the past policies is utilized to improve the reinforcement learning. This relies on the past errors exploiting the past policies. The Reuse Policy improves the speed of convergence of the learning process over the strategies that learn without reuse and combined with the use of the reinforcement learning, there is a 25-fold improvement in training. Experimentation is carried out on a database of 4750 samples. The overall recognition rate is found to be 90.65%.

Patent
15 Mar 2007
TL;DR: In this paper, the authors described techniques for automatic generation of one or more tags associated with an image file using hand-written annotations for a displayed image and handwriting recognition processing of the ink annotations.
Abstract: Techniques are described for performing automatic generation of one or more tags associated with an image file. One or more ink annotations for a displayed image are received. Handwriting recognition processing of the one or more ink annotations is performed. A string is generated and the string includes one or more recognized words used to form the one or more tags associated with the image file. The handwriting recognition processing and generating the string are performed in response to receiving the ink annotations.

Proceedings ArticleDOI
23 Sep 2007
TL;DR: This work aims at studying random forest methods in a strictly pragmatic approach, in order to provide rules on parameter settings for practitioners and draws some conclusions on random forest global behavior according to their parameter tuning.
Abstract: In the pattern recognition field, growing interest has been shown in recent years for multiple classifier systems and particularly for bagging, boosting and random sub-spaces. Those methods aim at inducing an ensemble of classifiers by producing diversity at different levels. Following this principle, Breiman has introduced in 2001 another family of methods called random forest. Our work aims at studying those methods in a strictly pragmatic approach, in order to provide rules on parameter settings for practitioners. For that purpose we have experimented the forest-RI algorithm, considered as the random forest reference method, on the MNIST handwritten digits database. In this paper, we describe random forest principles and review some methods proposed in the literature. We present next our experimental protocol and results. We finally draw some conclusions on random forest global behavior according to their parameter tuning.

Journal ArticleDOI
J.A. Pittman1
TL;DR: High-end versions of Microsoft's Vista now include tablet PC software, with an improved recognizer that supports both personalization and error reporting, and a time-delay neural network working with a lexicon.
Abstract: To support a wide range of writing styles and poorly formed cursive script, the Tablet PC input panel uses a time-delay neural network working with a lexicon. High-end versions of Microsoft's Vista now include tablet PC software, with an improved recognizer that supports both personalization and error reporting.

Patent
21 Dec 2007
TL;DR: In this paper, a character correction user interface allows a user to make corrections on an individual character basis and also provides other correction options for the word, allowing the user to edit a word of recognized text inline with other text by selecting the word.
Abstract: As a user writes using a handheld writing device, such as an electronic pen or stylus, handwriting input is received and initially displayed as digital ink The display of the digital ink is converted to recognized text inline with additional digital ink as the user continues to write A user may edit a word of recognized text inline with other text by selecting the word An enlarged version of the word is displayed in a character correction user interface that allows a user to make corrections on an individual character basis and also provides other correction options for the word

Journal ArticleDOI
TL;DR: A survey and an assessment of relevant papers appearing in recent publications of relevant conferences and journals are presented, identifying technical approaches that show promise in these areas as well as identifying the leading researchers for the applicable topics.
Abstract: Offline Chinese handwriting recognition (OCHR) is a typically difficult pattern recognition problem. Many authors have presented various approaches to recognizing its different aspects. We present a survey and an assessment of relevant papers appearing in recent publications of relevant conferences and journals, including those appearing in ICDAR, SDIUT, IWFHR, ICPR, PAMI, PR, PRL, SPIEDRR, and IJDAR. The methods are assessed in the sense that we document their technical approaches, strengths, and weaknesses, as well as the data sets on which they were reportedly tested and on which results were generated. We also identify a list of technology gaps with respect to Chinese handwriting recognition and identify technical approaches that show promise in these areas as well as identify the leading researchers for the applicable topics, discussing difficulties associated with any given approach.

Journal ArticleDOI
TL;DR: A new approach for online handwritten shape recognition that can deal with two-dimensional graphical shapes such as Latin and Asian characters, command gestures, symbols, small drawings, and geometric shapes is investigated.
Abstract: We investigate a new approach for online handwritten shape recognition. Interesting features of this approach include learning without manual tuning, learning from very few training samples, incremental learning of characters, and adaptation to the user-specific needs. The proposed system can deal with two-dimensional graphical shapes such as Latin and Asian characters, command gestures, symbols, small drawings, and geometric shapes. It can be used as a building block for a series of recognition tasks with many applications

Proceedings ArticleDOI
20 Oct 2007
TL;DR: In this paper, a new preprocessing technique for online handwriting is proposed, which removes the hooks of the strokes by using changed-angle threshold with length threshold, then filter the noise by using a smoothing technique, which is the combination of the cubic spline and the equal interpolation methods.
Abstract: This paper proposes a new preprocessing technique for online handwriting. The approach is to first remove the hooks of the strokes by using changed-angle threshold with length threshold, then filter the noise by using a smoothing technique, which is the combination of the cubic spline and the equal-interpolation methods. Finally, the handwriting is normalised. Experiments are carried out using the benchmark UNIPEN database. The experimental results show that our preprocessing technique can improve the recognition rates by at least 10%.

Proceedings ArticleDOI
23 Sep 2007
TL;DR: This work presents an effective method for writer identification in handwritten documents, developed a local approach, based on the extraction of characteristics that are specific to a writer, by a Bayesian classifier.
Abstract: This work presents an effective method for writer identification in handwritten documents. We have developed a local approach, based on the extraction of characteristics that are specific to a writer. To exploit the existence of redundant patterns within a handwriting, the writing is divided into a large number of small sub-images, and the sub-images that are morphologically similar are grouped together in the same classes. The patterns, which occur frequently for a writer are thus extracted. The author of the unknown document is then identified by a Bayesian classifier. The system trained and tested on 50 documents of the same number of authors, reported an identification rate of 94%.

Proceedings ArticleDOI
10 Sep 2007
TL;DR: A brief overview of the field of handwritten signature verification is presented and some of the most relevant perspectives are highlighted.
Abstract: In the information and communication personal verification is a crucial aspect. Among the different means for personal verification, handwritten signature plays a fundamental role since signature is the most diffuse means for personal verification in our daily life. In this paper a brief overview of the field of handwritten signature verification is presented and some of the most relevant perspectives are highlighted.

Journal ArticleDOI
TL;DR: The geometric and statistical features used in the recognizer and the all-pairs classification algorithm are described and the results of experiments are presented that quantify the effect incorporating a writer-independent recognition engine into aWriter-dependent recognizer has on accuracy, speed, and user training time.
Abstract: We present a practical technique for using a writer-independent recognition engine to improve the accuracy and speed while reducing the training requirements of a writer-dependent symbol recognizer. Our writer-dependent recognizer uses a set of binary classifiers based on the AdaBoost learning algorithm, one for each possible pairwise symbol comparison. Each classifier consists of a set of weak learners, one of which is based on a writer-independent handwriting recognizer. During online recognition, we also use the n-best list of the writer-independent recognizer to prune the set of possible symbols and, thus, reduce the number of required binary classifications. In this paper, we describe the geometric and statistical features used in our recognizer and our all-pairs classification algorithm. We also present the results of experiments that quantify the effect incorporating a writer-independent recognition engine into a writer-dependent recognizer has on accuracy, speed, and user training time.

Proceedings ArticleDOI
02 Apr 2007
TL;DR: In this paper, a modified exponential membership function fitted to the fuzzy sets derived from normalized distance features obtained using the box approach was used for Hindi numerals recognition, and the overall recognition rate was found to be 95% with the use of reinforcement learning.
Abstract: This paper presents the recognition of handwritten Hindi numerals based on the modified exponential membership function fitted to the fuzzy sets derived from normalized distance features obtained using the box approach. The exponential membership function is modified by two structural parameters that are estimated by optimizing the criterion function associated with the input fuzzy modeling. We then utilize a `reuse policy' that provides guidance from past error values of the criteria function to accomplish the reinforcement learning. We also show how the `reuse policy' improves the speed of convergence of the learning process over other strategies that learn without reuse. There is a 25-fold improvement in training with the use of the reinforcement learning. Experimentation is carried out on a limited database of around 3500 Hindi numeral samples. The overall recognition rate is found to be 95%

Proceedings ArticleDOI
01 Feb 2007
TL;DR: The IFN/ENIT-database of handwritten Tunisian town names is used by many research groups working on recognition systems and an example of using the data for developing baseline estimation methods is given.
Abstract: Databases enclosing a huge amount of images of handwritten words together with detailed ground truth information are the most important precondition for the development of handwritten word recognition systems. The IFN/ENIT-database of handwritten Tunisian town names is used by many research groups working on recognition systems. This paper gives at first a short overview about the most important features of the IFN/ENIT-database. In the second part an example of using the data for developing baseline estimation methods is given. In the third part a recognition system is described and some results are shown.

Proceedings ArticleDOI
17 Jun 2007
TL;DR: This paper proposes a novel framework for offline signature verification which makes use of online handwriting instead of handwritten images for registration and develops a verification criterion which combines the duration and amplitude variances of handwriting.
Abstract: This paper proposes a novel framework for offline signature verification. Different from previous methods, our approach makes use of online handwriting instead of handwritten images for registration. The online registrations enable robust recovery of the writing trajectory from an input offline signature and thus allow effective shape matching between registration and verification signatures. In addition, we propose several new techniques to improve the performance of the new signature verification system: 1. we formulate and solve the recovery of writing trajectory within the framework of conditional random fields; 2. we propose a new shape descriptor, online context, for aligning signatures; 3. we develop a verification criterion which combines the duration and amplitude variances of handwriting. Experiments on a benchmark database show that the proposed method significantly outperforms the well-known offline signature verification methods and achieve comparable performance with online signature verification methods.

Proceedings ArticleDOI
23 Sep 2007
TL;DR: A new interactive, on-line framework which, rather than full automation, aims at assisting the human in the proper recognition- transcription process; that is, facilitate and speed up their transcription task of handwritten texts.
Abstract: To date, automatic handwriting recognition systems are far from being perfect and often they need a post editing where a human intervention is required to check and correct the results of such systems. We propose to have a new interactive, on-line framework which, rather than full automation, aims at assisting the human in the proper recognition- transcription process; that is, facilitate and speed up their transcription task of handwritten texts. This framework combines the efficiency of automatic handwriting recognition systems with the accuracy of the human transcriptor. The best result is a cost-effective perfect transcription of the handwriting text images.

Proceedings ArticleDOI
23 Sep 2007
TL;DR: It is shown that using unsupervised learning to initialize the layers of a deep network dramatically reduces the required number of training samples, particularly for such tasks as the recognition of everyday objects at the category level.
Abstract: The machine learning and pattern recognition communities are facing two challenges: solving the normalization problem, and solving the deep learning problem. The normalization problem is related to the difficulty of training probabilistic models over large spaces while keeping them properly normalized. In recent years, the ML and natural language communities have devoted considerable efforts to circumventing this problem by developing "un-normalized" learning models for tasks in which the output is highly structured (e.g. English sentences). This class of models was in fact originally developed during the 90's in the handwriting recognition community, and includes graph transformer networks, conditional random fields, hidden Markov SVMs, and maximum margin Markov networks. We describe these models within the unifying framework of "energy-based models" (EBM). The deep learning problem is related to the issue of training all the levels of a recognition system (e.g. segmentation, feature extraction, recognition, etc) in an integrated fashion. We first consider " traditional" methods for deep learning, such as convolutional networks and back-propagation, and show that, although they produce very low error rates for handwriting and object recognition, they require many training samples. We show that using unsupervised learning to initialize the layers of a deep network dramatically reduces the required number of training samples, particularly for such tasks as the recognition of everyday objects at the category level.

Proceedings ArticleDOI
23 Sep 2007
TL;DR: In experiments on the TUAT Kon-date database, the proposed MRF approach yield superior performance compared to individual stroke classification and sequence classification based on hidden Markov models (HMMs).
Abstract: In this paper, we present an approach for separating text and non-text ink strokes in online handwritten Japanese documents based on Markov random fields (MRFs), which effectively utilize the spatial relationship between strokes. Support vector machine (SVM) classifiers are trained for individual stroke and stroke pair classification, and on converting the SVM outputs to probabilities, the likelihood clique potentials of MRF are derived. In experiments on the TUAT Kon-date database, the proposed MRF approach yield superior performance compared to individual stroke classification and sequence classification based on hidden Markov models (HMMs).