Showing papers on "Handwriting recognition published in 2006"

Offline Arabic handwriting recognition: a survey

Abstract: The automatic recognition of text on scanned images has enabled many applications such as searching for words in large volumes of documents, automatic sorting of postal mail, and convenient editing of previously printed documents. The domain of handwriting in the Arabic script presents unique technical challenges and has been addressed more recently than other domains. Many different methods have been proposed and applied to various types of images. This paper provides a comprehensive review of these methods. It is the first survey to focus on Arabic handwriting recognition and the first Arabic character recognition survey to provide recognition rates and descriptions of test data for the approaches discussed. It includes background on the field, discussion of the methods, and future research directions.

“One Against One” or “One Against All”: Which One is Better for Handwriting Recognition with SVMs?

Abstract: The “one against one” and the “one against all” are the two most popular strategies for multi-class SVM; however, according to the literature review, it seems impossible to conclude which one is better for handwriting recognition. Thus, we compared these two classical strategies on two different handwritten character recognition problems. Several post-processing methods for estimating posterior probability were also evaluated and the results were compared with the ones obtained using MLP. Finally, the “one against all” strategy appears significantly more accurate for digit recognition, while the difference between the two strategies is much less obvious with upper-case letters. Besides, the “one against one” strategy is substantially faster to train and seems preferable for problems with a very large number of classes. To conclude, SVMs allow significantly better estimation of probabilities than MLP, which is promising from the point of view of their incorporation into handwriting recognition systems.

Online Arabic Handwriting Recognition Using Hidden Markov Models

Abstract: Online handwriting recognition of Arabic script is a difficult problem since it is naturally both cursive and unconstrained. The analysis of Arabic script is further complicated in comparison to Latin script due to obligatory dots/stokes that are placed above or below most letters. This paper introduces a Hidden Markov Model (HMM) based system to provide solutions for most of the difficulties inherent in recognizing Arabic script including: letter connectivity, position-dependent letter shaping, and delayed strokes. This is the first HMM-based solution to online Arabic handwriting recognition. We report successful results for writerdependent and writer-independent word recognition.

HMM-Based On-Line Recognition of Handwritten Whiteboard Notes

Abstract: In this paper we present an on-line recognition system for handwritten texts acquired from a whiteboard. This input modality has received relatively little attention in the handwriting recognition community in the past. The system proposed in this paper uses state-of-the-art normalization and feature extraction strategies to transform a handwritten text line into a sequence of feature vectors. Additional preprocessing techniques are introduced, which significantly increase the word recognition rate. For classification, Hidden Markov Models are used together with a statistical language model. In writer independent experiments we achieved word recognition rates of 67.3% on the test set when no language model is used, and 70.8% by including a language model.

Personalized handwriting recognition via biased regularization

Abstract: We present a new approach to personalized handwriting recognition. The problem, also known as writer adaptation, consists of converting a generic (user-independent) recognizer into a personalized (user-dependent) one, which has an improved recognition rate for a particular user. The adaptation step usually involves user-specific samples, which leads to the fundamental question of how to fuse this new information with that captured by the generic recognizer. We propose adapting the recognizer by minimizing a regularized risk functional (a modified SVM) where the prior knowledge from the generic recognizer enters through a modified regularization term. The result is a simple personalization framework with very good practical properties. Experiments on a 100 class real-world data set show that the number of errors can be reduced by over 40% with as few as five user samples per character.

CueTIP: a mixed-initiative interface for correcting handwriting errors

Abstract: With advances in pen-based computing devices, handwriting has become an increasingly popular input modality. Researchers have put considerable effort into building intelligent recognition systems that can translate handwriting to text with increasing accuracy. However, handwritten input is inherently ambiguous, and these systems will always make errors. Unfortunately, work on error recovery mechanisms has mainly focused on interface innovations that allow users to manually transform the erroneous recognition result into the intended one. In our work, we propose a mixed-initiative approach to error correction. We describe CueTIP, a novel correction interface that takes advantage of the recognizer to continually evolve its results using the additional information from user corrections. This significantly reduces the number of actions required to reach the intended result. We present a user study showing that CueTIP is more efficient and better preferred for correcting handwriting recognition errors. Grounded in the discussion of CueTIP, we also present design principles that may be applied to mixed-initiative correction interfaces in other domains.

Enhancing Binary Feature Vector Similarity Measures

Abstract: Similarity and dissimilarity measures play an important role in pattern classification and clustering. For a century, researchers have searched for a good measure. Here, we review, categorize, and evaluate various binary vector similarity / dissimilarity measures. One of the most contentious disputes in the similarity measure selection problem is whether the measure includes or excludes negative matches. While inner-product based similarity measures consider only positive matches, other conventional measures credit both positive and negative matches equally. Hence, we propose an enhanced similarity measure that gives variable credits and show that it is superior to conventional measures in IRIS biometric authentication and offline handwritten character recognition applications. Finally, the proposed similarity measure can be further boosted by applying weights and we demonstrate that it outperforms the weighted Hamming distance.

A New Algorithm for Detecting Text Line in Handwritten Documents

Abstract: Curvilinear text line detection and segmentation in handwritten documents is a significant challenge for handwriting recognition. Given no prior knowledge of script, we model text line detection as an image segmentation problem by enhancing text line structure using a Gaussian window, and adopting the level set method to evolve text line boundaries. Experiments show that the proposed method achieves high accuracy for detecting text lines in both handwritten and machine printed documents with many scripts.

Handwritten Signature Verification Using Image Invariants and Dynamic Features

Abstract: In this paper, a development of automatic signature classification system is proposed. We have presented offline and online signature verification system, based on the signature invariants and its dynamic features. The proposed system segments each signature based on its perceptually important points and then, for each segment, computes a number of features that are scale, rotation and displacement invariant. The normalized moments and the normalized Fourier descriptors are used for this invariancy, while the speed of pen is used as a dynamic feature of the signature. In both cases the data acquisition, pre-processing, feature extraction and comparison steps are analyzed and discussed. Both static and dynamic features were used as an input to a neural network. The neural network used for classification is a multi-layer perception (MLP) with one input layer, one hidden layer and one output layer. The performance of the proposed system is presented through simulation examples.

Minimum classification error training for online handwriting recognition

Abstract: This paper describes an application of the minimum classification error (MCE) criterion to the problem of recognizing online unconstrained-style characters and words. We describe an HMM-based, character and word-level MCE training aimed at minimizing the character or word error rate while enabling flexibility in writing style through the use of multiple allographs per character. Experiments on a writer-independent character recognition task covering alpha-numerical characters and keyboard symbols show that the MCE criterion achieves more than 30 percent character error rate reduction compared to the baseline maximum likelihood-based system. Word recognition results, on vocabularies of 5k to 10k, show that MCE training achieves around 17 percent word error rate reduction when compared to the baseline maximum likelihood system

Recognition of Handwritten Kannada Numerals

Abstract: This paper deals with a quadratic classifier based scheme for the recognition of off-line handwritten numerals of Kannada, an important Indian script. The features used in the classifier are obtained from the directional chain code information of the contour points of the characters. The bounding box of a character is segmented into blocks and the chain code histogram is computed in each of the blocks. Here we have used 64 dimensional and 100 dimensional features for a comparative study on the recognition accuracy of our proposed system. This chain code features are fed to the quadratic classifier for recognition. We tested our scheme on 2300 data samples and obtained 97.87% and 98.45% recognition accuracy using 64 dimensional and 100 dimensional features respectively, from the proposed scheme using five-fold cross-validation technique.

Searching Off-line Arabic Documents

Abstract: Currently an abundance of historical manuscripts, journals, and scientific notes remain largely unaccessible in library archives. Manual transcription and publication of such documents is unlikely, and automatic transcription with high enough accuracy to support a traditional text search is difficult. In this work we describe a lexicon-free system for performing text queries on off-line printed and handwritten Arabic documents. Our segmentation-based approach utilizes gHMMs with a bigram letter transition model, and KPCA/LDA for letter discrimination. The segmentation stage is integrated with inference. We show that our method is robust to varying letter forms, ligatures, and overlaps. Additionally, we find that ignoring letters beyond the adjoining neighbors has little effect on inference and localization, which leads to a significant performance increase over standard dynamic programming. Finally, we discuss an extension to perform batch searches of large word lists for indexing purposes.

Offline grammar-based recognition of handwritten sentences

Abstract: This paper proposes a sequential coupling of a hidden Markov model (HMM) recognizer for offline handwritten English sentences with a probabilistic bottom-up chart parser using stochastic context-free grammars (SCFG) extracted from a text corpus. Based on extensive experiments, we conclude that syntax analysis helps to improve recognition rates significantly.

Multi-lingual offline handwriting recognition using hidden Markov models: a script-independent approach

Abstract: This paper introduces a script-independent methodology for multilingual offline handwriting recognition (OHR) based on the use of Hidden Markov Models (HMM). The OHR methodology extends our script-independent approach for OCR of machine-printed text images. The feature extraction, training, and recognition components of the system are all designed to be script independent. The HMM training and recognition components are based on our Byblos continuous speech recognition system. The HMM parameters are estimated automatically from the training data, without the need for laborious hand-written rules. The system does not require pre-segmentation of the data, neither at the word level nor at the character level. Thus, the system can handle languages with cursive handwritten scripts in a straightforward manner. The script independence of the system is demonstrated with experimental results in three scripts that exhibit significant differences in glyph characteristics: English, Chinese, and Arabic. Results from an initial set of experiments are presented to demonstrate the viability of the proposed methodology.

Cursive character handwriting recognition system and method

Abstract: A cursive character handwriting recognition system includes image processing means for processing an image of a handwritten word of one or more characters and classification means for determining an optimal string of one or more characters as composing the imaged word. The processing means segments the characters such that each character is made up of one or more segments and determines a sequence of the segments using an over-segmentation-relabeling algorithm. The system also includes feature extraction means for deriving a feature vector to represent feature information of one segment or a combination of several consecutive segments. The over-segmentation-relabeling algorithm places certain segments considered as diacritics or small segments so as to immediately precede or follow a segment of the associated main character body. Additionally, the system also includes classification means that processes each string of segments and outputs a number of optimal strings which could be matched against a given lexicon.

Feature selection for ensembles applied to handwriting recognition

Abstract: Feature selection for ensembles has shown to be an effective strategy for ensemble creation due to its ability of producing good subsets of features, which make the classifiers of the ensemble disagree on difficult cases. In this paper we present an ensemble feature selection approach based on a hierarchical multi-objective genetic algorithm. The underpinning paradigm is the "overproduce and choose". The algorithm operates in two levels. Firstly, it performs feature selection in order to generate a set of classifiers and then it chooses the best team of classifiers. In order to show its robustness, the method is evaluated in two different contexts:supervised and unsupervised feature selection. In the former, we have considered the problem of handwritten digit recognition and used three different feature sets and multi-layer perceptron neural networks as classifiers. In the latter, we took into account the problem of handwritten month word recognition and used three different feature sets and hidden Markov models as classifiers. Experiments and comparisons with classical methods, such as Bagging and Boosting, demonstrated that the proposed methodology brings compelling improvements when classifiers have to work with very low error rates. Comparisons have been done by considering the recognition rates only.

Exploring the use of conditional random field models and HMMs for historical handwritten document recognition

Abstract: In this paper we explore different approaches for improving the performance of dependency models on discrete features for handwriting recognition. Hidden Markov models have often been used for handwriting recognition. Conditional random fields (CRF's) allow for more general dependencies and we investigate their use. We believe that this is the first attempt at apply CRF's for handwriting recognition. We show that on the whole word recognition task, the CRF performs better than a HMM on a publicly available standard dataset of 20 pages of George Washington's manuscripts. The scale space for the whole word recognition task is large - almost 1200 states. To make CRF computation tractable we use beam search to make inference more efficient using three different approaches. Better improvement can be obtained using the HMM by directly smoothing the discrete features using the collection frequencies. This shows the importance of smoothing and also indicates the difficulty of training CRF's when large state spaces are involved.

Databases and competitions: strategies to improve Arabic recognition systems

Abstract: The great success and high recognition rates of both OCR systems and recognition systems for handwritten words are unconceivable without the availability of huge datasets of real world data. This chapter gives a short survey of datasets used for recognition with special focus on their application. The main part of this chapter deals with Arabic handwriting, datasets for recognition systems, and their availability. A description of different datasets and their usability is given, and the results of a competition are presented. Finally, a strategy for the development of Arabic handwriting recognition systems based on datasets and competitions is presented.

Separating lines of text in free-form handwritten historical documents

Abstract: We present an approach to finding (and separating) lines of text in free-form handwritten historical document images. After preprocessing, our method uses the count of foreground/background transitions in a binarized image to determine areas of the document that are likely to be text lines. Alternatively, an adaptive local connectivity map (ALCM) found in the literature can be used for this step of the process. We then use a min-cut/max-flow graph cut algorithm to split up text areas that appear to encompass more than one line of text. After removing text lines containing relatively little text information (or merging them with nearby text lines), we create output images for each line. A grayscale output image is created, as well as a special mask image containing both the foreground and information flagging ambiguous pixels. Foreground pixels that belong to other text lines are removed from the output images to provide cleaner line images useful for further processing. While some refinement is still necessary, the result of early experimentation with our method is encouraging.

Off-lineWriter Identification Using Gaussian Mixture Models

Abstract: Writer identification is the task of determining the author of a sample handwriting from a set of writers In this paper, we propose Gaussian Mixture Models (GMMs) to address the task of off-line, text independent writer identification of text lines The resulting system is compared to a system that uses a Hidden Markov Model (HMM) based approach While the GMM based system is conceptually much simpler and faster to train than the HMM based system, it achieves a significantly higher writer identification rate of 9846% on a data set of 4,103 text lines coming from 100 writers

Novel Hybrid NN/HMM Modelling Techniques for On-line Handwriting Recognition

Abstract: In this work we propose two hybrid NN/HMM systems for handwriting recognition. The tied posterior model approximates the output probability density function of a Hidden Markov Model (HMM) with a neural net (NN). This allows a discriminative training of the model. The second system is the tandem approach: A NN is used as part of the feature extraction, and then a standard HMM apporach is applied. This adds more discrimination to the features. In an experimental section we compare the two proposed models with a baseline standard HMM system. We show that enhancing the feature vector has only a limited effect on the standard HMMs, but a significant influence to the hybrid systems. With an enhanced feature vector the two hybrid models highly outperform all baseline models. The tandem approach improves the recognition performance by 4.6% (52.9% rel. error reduction) absolute compared to the best baseline HMM.

Speech pen: predictive handwriting based on ambient multimodal recognition

Abstract: It is tedious to handwrite long passages of text by hand. To make this process more efficient, we propose predictive handwriting that provides input predictions when the user writes by hand. A predictive handwriting system presents possible next words as a list and allows the user to select one to skip manual writing. Since it is not clear if people are willing to use prediction, we first run a user study to compare handwriting and selecting from the list. The result shows that, in Japanese, people prefer to select, especially when the expected performance gain from using selection is large. Based on these observations, we designed a multimodal input system, called speech-pen, that assists digital writing during lectures or presentations with background speech and handwriting recognition. The system recognizes speech and handwriting in the background and provides the instructor with predictions for further writing. The speech-pen system also allows the sharing of context information for predictions among the instructor and the audience; the result of the instructor's speech recognition is sent to the audience to support their own note-taking. Our preliminary study shows the effectiveness of this system and the implications for further improvements.

Hidden Markov models combining discrete symbols and continuous attributes in handwriting recognition

Abstract: Prior arts in handwritten word recognition model either discrete features or continuous features, but not both. This paper combines discrete symbols and continuous attributes into structural handwriting features and model, them by transition-emitting and state-emitting hidden Markov models. The models are rigorously defined and experiments have proven their effectiveness.

System and method for automated reading of handwriting

Abstract: A system for automatically recognizing a handwriting image and converting such image to text data including a sequence of validated words, has an image input device, a number of handwriting recognition engines, and control unit. A first handwriting recognition engine is responsive to the image input device, for analyzing the data file and providing one or more possible text words for each successive word in the data file. The first handwriting recognition engine further provides a resemblance indication for each possible text word indicating a level of resemblance between its appearance and the appearance of the handwritten word in the data file. In the event that there is not a high level of confidence in the selection of the first handwriting recognition engine, a selection of a validated word is based on the selections of one or more of the other handwriting recognition engines.

A Complete Handwritten Numeral Database of Bangla – A Major Indic Script

Abstract: This paper describes the ISI database of handwritten Bangla numerals Bangla is the second most popular language and script of the Indian subcontinent and it is used by more than 200 million people all over the globe The present database has several components which include both on-line and off-line handwritten numerals Samples of numeral strings and isolated numerals have been collected under both modes of writing This database has been developed at the Computer Vision and Pattern Recognition Unit laboratory of Indian Statistical Institute, Kolkata Samples of the present database are properly ground thruthed and subdivided into respective training and test sets The off-line sample images are stored in TIFF image format and the on-line samples are stored along with various information as header in ASCII file format This database will facilitate fruitful research on handwriting recognition of Bangla through free access to the researchers

A Hybrid HMM-SVM Method for Online Handwriting Symbol Recognition

Abstract: This paper presents a combined approach for online handwriting symbols recognition. The basic idea of this approach is to employ a set of left-right HMMs as a feature extractor to produce HMM features, and combine them with global features into a new feature vector as input, and then use SVM as a classifier to finally identify unknown symbols. The new feature vector consists of the global features and several pairs of maximum probabilities with their associated different model labels. A recogniser based on this method inherits the practical and dynamical modeling abilities from HMM, and robust discriminating ability from SVM for classification tasks. This technique also reduces the dimensions of feature vectors significantly and solves the speed and size problem when using only SVM. The experimental results show that this combined hybrid approach outperforms several classifiers reported in recent researches, and could achieve recognition rates of 97.48%, 91.99% and 91.74% for digits and upper/lower case characters respectively on the UNIPEN database benchmarks.