By Topic

Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on

Date 22-22 Sept. 1999

Filter Results

Displaying Results 1 - 25 of 203
  • Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318)

    Save to Project icon | Request Permissions | PDF file iconPDF (528 KB)  
    Freely Available from IEEE
  • Handwritten numeral string recognition with stroke grouping

    Page(s): 745 - 748
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (60 KB)  

    In this paper a framework for off-line handwritten numeral string recognition based on stroke grouping is proposed. In our approach, strokes are aligned into a sequence of strokes and then a segmentation process is performed to partition strokes in the sequence into possible-digits, that is, groups of strokes which may be a digit. As a result of stroke grouping, grouping-hypotheses, which imply possible segmentation, are generated. An input numeral string is recognized by a dynamic programming scheme, in which the best grouping-hypothesis with maximum matching score is chosen. The framework also provides a systematic way of reducing computational complexity by embedding external knowledge into the framework. The experimental results to evaluate the proposed framework are shown. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Handwritten numeral recognition by means of evolutionary algorithms

    Page(s): 804 - 807
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (71 KB)  

    We present a handwritten numeral recognition system centered on a novel method for extracting the set of prototypes to be used during the classification. The method is based on an evolutionary learning mechanism that exploits a genetic algorithm with niching for producing the best set of prototypes. By combining the search power of genetic algorithms and the ability of niching mechanisms to maintain different prototypes during the evolution, the proposed method allows to obtain as many prototypes as needed to model the variability exhibited by the samples belonging to each class. Such a learning mechanism overcomes the limitations of other evolutionary learning methods proposed in the literature for dealing with problems characterized by a large amount of variability in the data set as in the case of handwriting recognition. Experiments have proved that the performance of the system is comparable with, or even better than that exhibited by a neural classifier. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • Index of authors

    Page(s): 817 - 821
    Save to Project icon | Request Permissions | PDF file iconPDF (17 KB)  
    Freely Available from IEEE
  • EXTRAFOR: automatic EXTRAction of mathematical FORmulas

    Page(s): 527 - 530
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (64 KB)  

    We present a method for automatic extraction of mathematical formulas from images of documents without character recognition. Formula extraction is first done by location of its most significant symbols, then extension to adjoining symbols using contextual rules until delimitation of the whole formula space. Mathematical symbol labelling is realised from models created at the learning stage using fuzzy logic. From the experiments, we found that the average rate of primary labelling of mathematical symbols is about 95.3%. The obtained results have demonstrated the applicability of our system since 90% of mathematical formulas are well extracted from documents printed with high quality View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recovery of drawing order from scanned images of multi-stroke handwriting

    Page(s): 261 - 264
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (72 KB)  

    Describes a method to recover the drawing order of a multi-stroke handwritten script from a binary 2D image. First, we construct a graph from the scanned image by applying a thinning process and extracting the skeletal pixels. Next, we identify the start or end vertices in the graph, and then globally analyze the graph to label it by determining the types of each vertex and each edge. Finally, we trace all the strokes using the labeling information and recover the drawing order. The method does not enumerate the possible drawing orders and does not cause a combinatorial explosion, even if the script is very complex. By recovering the drawing order of a handwritten script, the temporal information can be recovered from a scanned image. Hence, this method can be used as a bridge from the offline handwriting character recognition problem to the online one View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Structured document labeling and rule extraction using a new recurrent fuzzy-neural system

    Page(s): 181 - 184
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (56 KB)  

    Proposes an approach to the problem of logical labeling in structured documents, employing a new recurrent neuro-fuzzy system called RFasArt (Recurrent Fuzzy Adaptive System, ART-based). RFasArt preserves the fine characteristics, such as modularity, stability and flexibility, of its predecessors Fuzzy ARTMAP and FasArt. In this paper, the documents are considered as pseudo-temporal sequences, and context information is exploited in an integrated form. Two working prototypes for a MIME-based mailing system and for a digital library were tested with over 90% of the recognition rate and less ambiguous decisions than in the previous systems. A manageable knowledge base was constructed using fuzzy rules that were easily interpretable by human users, and examples of rule creation and fusion are shown View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generalised projections: a tool for cursive handwriting normalisation

    Page(s): 729 - 732
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (136 KB)  

    The paper addresses the problem of removing unwanted variations in cursive handwriting by estimating and compensating reference lines and slant angle. To this purpose, the paper introduces the mathematical framework of generalized projections (GP), an extension of the well known projection profiles (PP). This framework proved to be quite effective and robust in the estimation of those parameters. The particular application of GP to skew angle estimation, slant estimation and reference lines detection is then described. Experimental results are finally presented and discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dictionary preselection in a neuro-Markovian word recognition system

    Page(s): 539 - 542
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (32 KB)  

    Previously we have introduced a neural predictive system for on-line and off-line word recognition (Garcia-Salicetti et al., 1995; 1996; 1997). Words are recognized thanks to a dictionary which is used in a postprocessing stage. We focus on this lexical part of the system and more precisely on the preselection technique that is used to reduce the computational complexity. We define and compare two edit distances and show how the results can be improved through the use of the confusion matrix View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Quality evaluation of document segmentation results

    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (12 KB)  

    Summary form only given, as follows. Increasing the performance of document analysis systems requires a detailed quality evaluation of the achieved results. By focussing on segmentation algorithms, we point out that the results produced from the module under consideration should be evaluated directly; we show that the text based evaluation method which is often used in the document analysis domain is not sufficient for the purpose of a detailed quality evaluation of the segmentation module. Therefore, we propose a general evaluation approach for comparing segmentation results which is based on the segments directly. This approach is able to handle both algorithms which produce complete segmentations (partition) and algorithms which only extract objects of interest (extraction). Classes of errors are defined in a systematic way and frequencies for each class can be computed. The evaluation approach is applicable to segmentation or extraction algorithms in a wide range. We have chosen the character segmentation task as an example to demonstrate the applicability of our evaluation approach and we suggest applying our approach to other segmentation tasks View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feature extraction by fractal dimensions

    Page(s): 217 - 220
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (100 KB)  

    Proposes a method that reduces the dimensionality of a 2D pattern by means of a central projection approach, and thereafter performs a Daubechies wavelet transformation on the derived 1D pattern to generate a set of wavelet transformation sub-patterns, namely curves that are non-self-intersecting. Further, from the resulting non-self-intersecting curves, the divider dimensions are compared with the modified box-counting approach. These divider dimensions constitute a new feature vector for the original 2D pattern, defined over the curve's fractal dimensions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic classification of deformed handwritten numeral characters

    Page(s): 269 - 272
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (48 KB)  

    Describes a method which utilizes Hopfield neural nets to classify those handwritten numerals presenting deformations and stylistic traces. Information for the classification consists of some topological image features and the image pixel distribution. If the recognition cannot be done by these features due to noise and deformations in the images of the numerals, the classification process is performed by four Hopfield neural nets. Using four such nets, we are able to minimize the problem caused by correlated patterns, and also to increase the neural classifier's pattern storage capacity. The proposed method was tested on 121 Brazilian bank checks, achieving a 92.4% correct recognition rate View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Document analysis in gray level and typography extraction using character pattern redundancies

    Page(s): 177 - 180
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (200 KB)  

    The paper describes the processing of magazine or newspaper images which need to be segmented in gray level. The first part proposes an original method to extract the physical layout of gray-level documents. The second part of the paper describes a rough logical structure by analyzing the typography, aiming to extract relevant information about the logical layout by combining information about colors, typography, and the physical structural layout for use by an automatic document indexation system. Character prototypes were automatically extracted by grouping characters which have the same binary patterns. We suggest using this character-grouping method to extract typographical information and recognize different font styles and sizes used in the document View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Methodology for flexible and efficient analysis of the performance of page segmentation algorithms

    Page(s): 451 - 454
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (48 KB)  

    The paper presents part of a new DIA performance analysis framework aimed at layout analysis algorithm developers. A new region representation scheme (an interval based description of isothetic polygons) and a corresponding comparison approach are introduced. These enable fast and accurate geometric comparison of ground truth with results of page segmentation, improving on current evaluation methods. Complex layouts are accurately described and layout analysis methods that handle them can be effectively evaluated. A further benefit of the new approach is that it measures the accuracy of the description of regions, an issue which is important for complex layouts involving non text regions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A document classification and extraction system with learning ability

    Page(s): 197 - 200
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (40 KB)  

    Document image processing begins at the OCR phase with the difficulty of automatic document analysis and understanding. Most existing systems only do well in their specific application domains. In this paper, we describe a domain-independent automatic document image understanding system with learning ability. A segmentation method based on “logical closeness” is proposed. A novel and natural representation of document layout structure-a directed weight graph (DWG)-is described. To classify a given document, a string representation matching algorithm is applied first, instead of comparing all the sample graphs. A frame template and a document type hierarchy (DTH) are used to represent the document's logical structure and the hierarchical relationships among these frame templates, respectively. In this paper, two learning methodologies are applied-learning from experience and an enhanced perceptron learning algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Whole word recognition in facsimile images

    Page(s): 547 - 550
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (36 KB)  

    This paper presents the research carried out in producing a whole recognizor for cursive handwritten words in facsimile images. Two sets of handwritten data samples are collected and converted into facsimile images. The first set comprises approximately 1600 word images from 8 writers and is used for development purposes. The second set consists of approximately 2000 word images from 10 writers. This set is used for testing only. The algorithms for extraction of holistic features namely, vertical bars, holes and cups used in the recognizor are described. A series of test are carried out and the results are presented using a 200 word lexicon. The holistic recognizor produced 62% top rank and 82% in top 5 alternatives. When a lexicon of 1000 words was used these values reduced to 49% and 70% respectively. The future directions of the research for improvement of recognition rate are proposed. It is envisaged that definition of further features would improve the overall accuracy View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A method for street number matching in Japanese address recognition

    Page(s): 321 - 324
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (64 KB)  

    A method for street number matching in Japanese address recognition is presented. This method uses the knowledge about the number expression formats, which is represented by the expression patterns of street numbers. The expression patterns are composed of character class symbols which represent numerals and delimiter characters. By matching the expression patterns against the character candidates in the lattice which is the result of character recognition, the class symbol lattice is generated. In the class symbol lattice, the numeral and delimiter character candidates are replaced by the corresponding class symbol. An experiment using 6,136 expression patterns showed that 93% of street numbers on the test mail-pieces were correctly recognized View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Handwritten character recognition using monotonic and continuous two-dimensional warping

    Page(s): 499 - 502
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (276 KB)  

    In this paper, a handwritten character recognition experiment using a monotonic and continuous two-dimensional warping algorithm is reported. This warping algorithm is based on dynamic programming and searches for the optimal pixel-to-pixel mapping between given two images subject to two-dimensional monotonicity and continuity constraints. Experimental comparisons with rigid matching and local perturbation show the performance superiority of the monotonic and continuous warping in character recognition View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extraction of personal features from stroke shape, writing pressure and pen inclination in ordinary characters

    Page(s): 426 - 429
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (192 KB)  

    We propose an extraction method of personal features based on online handwritten characters including writing pressure and pen inclination information. In the proposed method, any handwritten character (i.e., ordinary character) is described by a set of three dimensional curves, and personal features are described by a set of Fourier descriptors for the three dimensional curves. From some simulation results using handwritten data, it is clear that the proposed method effectively extracts personal features from ordinary characters View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Information theoretic analysis of postal address fields for automatic address interpretation

    Page(s): 309 - 312
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (80 KB)  

    This paper concerns a study of information content in postal address fields for automatic address interpretation. Information provided by a combination of address components and information interaction among components is characterized in terms of Shannon's entropy. The efficiency of assignment strategies for determining a delivery point code can be compared by the propagation of uncertainty in address components. The quantity of redundancy between components can be computed from the information provided by these components. This information is useful in developing a strategy for selecting a useful component for recovering the value of an uncertain component. The uncertainty of a component based on another known component can be measured by conditional entropy. By ranking the uncertainty quantity, the effective processing flow for determining the value of a candidate component can be constructed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On-line adaptation in recognition of handwritten alphanumeric characters

    Page(s): 792 - 795
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (48 KB)  

    We have developed an adaptive online recognizer that is suitable for recognizing isolated alphanumeric characters. It is based on the k nearest neighbor rule. Various dissimilarity measures, all based on dynamic time warping (DTW), have been studied. The main focus of this work is on online adaptation. The adaptation is performed by modifying the prototype set of the classifier according to its recognition performance and the user's writing style. These adaptations include: (1) adding new prototypes, (2) inactivating confusing prototypes, and (3) reshaping existing prototypes. The reshaping algorithm is based on learning vector quantization (LVQ). The writers are allowed to use their own natural style of writing, and the adaptation is carried out during normal use in a self-supervised fashion and thus remains otherwise unnoticed by the user View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Handwritten numeral recognition using gradient and curvature of gray scale image

    Page(s): 277 - 280
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (260 KB)  

    Studies the use of curvature in addition to the gradient of gray-scale character images in order to improve the accuracy of handwritten numeral recognition. Three procedures, based on the curvature coefficient, biquadratic interpolation and gradient vector interpolation, are proposed for calculating the curvature of the equi-gray-scale curves of an input image. The efficiency of the feature vector is tested by recognition experiments for the handwritten numeral database IPTP CDROM1, which is a ZIP code database provided by the Institute for Posts and Telecommunications Policy (IPTP). The experimental results show the usefulness of the curvature feature, and a recognition rate of 99.40%, which is the highest that has ever been reported for this database, is achieved View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Region description and comparative analysis using a tesseral representation

    Page(s): 193 - 196
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (44 KB)  

    Presents a region representation scheme and comparative analysis methods based on a tesseral addressing system. The proposed scheme is described in the context of a performance analysis of page segmentation methods, a document image analysis area that is particularly sensitive to both a successful region description scheme and efficient methods for comparative analysis. The proposed tesseral representation is more economical in storage than other Cartesian-based approaches and can be advantageous for comparative analysis View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Writer adaptation of online handwriting models

    Page(s): 434 - 437
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (176 KB)  

    Writer adaptation is the process of converting a writer-independent handwriting recognition system, which models the characteristics of a large group of writers, into a writer-dependent system, which is tuned for a particular writer. Adaptation has the potential of increasing recognition accuracies, provided adequate models can be constructed for a particular writer. The limited amount of data that a writer typically provides makes the role of writer-independent models crucial in the adaptation process. Our approach to writer-adaptation makes use of writer-independent writing style models (called lexemes), to identify the styles present in a particular writer's training data. These models are then retrained using the writer's data. We demonstrate the feasibility of this approach using hidden Markov models trained on a combination of discretely and cursively written lower case characters. Our results show an average reduction in error rate of 16.3% for lower case characters as compared against representing each of the writer's character classes with a single model View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.