scispace - formally typeset
Search or ask a question

Showing papers on "Handwriting recognition published in 2005"


Patent
08 Feb 2005
TL;DR: In this paper, a hybrid approach to improve handwriting recognition and voice recognition in data process systems is described, where a front end is used to recognize strokes, characters and/or phonemes.
Abstract: A hybrid approach to improve handwriting recognition and voice recognition in data process systems is disclosed. In one embodiment, a front end is used to recognize strokes, characters and/or phonemes. The front end returns candidates with relative or absolute probabilities of matching to the input. Based on linguistic characteristics of the language, e.g. alphabetical or ideographic language for the words being entered, e.g. frequency of words and phrases being used, likely part of speech of the word entered, the morphology of the language, or the context in which the word is entered), a back end combines the candidates determined by the front end from inputs for words to match with known words and the probabilities of the use of such words in the current context.

251 citations


Proceedings ArticleDOI
31 Aug 2005
TL;DR: This paper describes the Arabic handwriting recognition competition held at ICDAR 2007, again uses the IFN/ENIT-database with Arabic handwritten Tunisian town names, and 8 groups with 14 systems are participating in the competition.
Abstract: This paper describes the Arabic handwriting recognition competition held at ICDAR 2007. This second competition (the first was at ICDAR 2005) again uses the IFN/ENIT-database with Arabic handwritten Tunisian town names. Today, more than 54 research groups from universities, research centers, and industry are working with this database worldwide. This year, 8 groups with 14 systems are participating in the competition. The systems were tested on known data and on two datasets which are unknown to the participants. The systems are compared on the most important characteristic, the recognition rate. Additionally, the relative speed of the different systems were compared. A short description of the participating groups, their systems, and the results achieved are finally presented.

220 citations


Patent
22 Mar 2005
TL;DR: In this article, a system for generating text responsive to both voice and handwriting input, including a microphone, stylus, and a tablet having an flat-panel display integrated into its surface, is presented.
Abstract: A system is disclosed for generating text responsive to both voice and handwriting input, including a microphone, stylus, and a tablet having an flat-panel display integrated into its surface. The system performs speech and handwriting recognition using a shared language model, which in one embodiment is trainable responsive to user correction of errors in either speech or handwriting recognition. Various other systems and methods are also disclosed.

215 citations


Proceedings ArticleDOI
31 Aug 2005
TL;DR: IAM-OnDB is a new large online handwritten sentences database that consists of text acquired via an electronic interface from a whiteboard and a recognizer for unconstrained English text that was trained and tested using this database.
Abstract: In this paper we present IAM-OnDB - a new large online handwritten sentences database. It is publicly available and consists of text acquired via an electronic interface from a whiteboard. The database contains about 86 K word instances from an 11 K dictionary written by more than 200 writers. We also describe a recognizer for unconstrained English text that was trained and tested using this database. This recognizer is based on hidden Markov models (HMMs). In our experiments we show that by using larger training sets we can significantly increase the word recognition rate. This recognizer may serve as a benchmark reference for future research.

208 citations


Journal ArticleDOI
TL;DR: A comparison of the two classifiers in off-line signature verification using random, simple and simulated forgeries to observe the capability of the classifiers to absorb intrapersonal variability and highlight interpersonal similarity.

199 citations


Journal ArticleDOI
TL;DR: A novel scale space algorithm for automatically segmenting handwritten (historical) documents into words is described and it is shown that the technique outperforms a state-of-the-art gap metrics word-segmentation algorithm on this collection.
Abstract: Many libraries, museums, and other organizations contain large collections of handwritten historical documents, for example, the papers of early presidents like George Washington at the Library of Congress. The first step in providing recognition/retrieval tools is to automatically segment handwritten pages into words. State of the art segmentation techniques like the gap metrics algorithm have been mostly developed and tested on highly constrained documents like bank checks and postal addresses. There has been little work on full handwritten pages and this work has usually involved testing on clean artificial documents created for the purpose of research. Historical manuscript images, on the other hand, contain a great deal of noise and are much more challenging. Here, a novel scale space algorithm for automatically segmenting handwritten (historical) documents into words is described. First, the page is cleaned to remove margins. This is followed by a gray-level projection profile algorithm for finding lines in images. Each line image is then filtered with an anisotropic Laplacian at several scales. This procedure produces blobs which correspond to portions of characters at small scales and to words at larger scales. Crucial to the algorithm is scale selection that is, finding the optimum scale at which blobs correspond to words. This is done by finding the maximum over scale of the extent or area of the blobs. This scale maximum is estimated using three different approaches. The blobs recovered at the optimum scale are then bounded with a rectangular box to recover the words. A post processing filtering step is performed to eliminate boxes of unusual size which are unlikely to correspond to words. The approach is tested on a number of different data sets and it is shown that, on 100 sampled documents from the George Washington corpus of handwritten document images, a total error rate of 17 percent is observed. The technique outperforms a state-of-the-art gap metrics word-segmentation algorithm on this collection.

199 citations


Journal ArticleDOI
TL;DR: This paper surveys the most significant problems in the area of offline document image processing, where connectionist-based approaches have been applied and depicts the most promising research guidelines in the field.
Abstract: Artificial neural networks have been extensively applied to document analysis and recognition. Most efforts have been devoted to the recognition of isolated handwritten and printed characters with widely recognized successful results. However, many other document processing tasks, like preprocessing, layout analysis, character segmentation, word recognition, and signature verification, have been effectively faced with very promising results. This paper surveys the most significant problems in the area of offline document image processing, where connectionist-based approaches have been applied. Similarities and differences between approaches belonging to different categories are discussed. A particular emphasis is given on the crucial role of prior knowledge for the conception of both appropriate architectures and learning algorithms. Finally, the paper provides a critical analysts on the reviewed approaches and depicts the most promising research guidelines in the field. In particular, a second generation of connectionist-based models are foreseen which are based on appropriate graphical representations of the learning environment.

196 citations


Proceedings ArticleDOI
31 Aug 2005
TL;DR: A 1D HMM offline handwriting recognition system employing an analytical approach supported by a set of robust language independent features extracted on binary images and learns character models without character pre-segmentation.
Abstract: In this paper, we describe a 1D HMM offline handwriting recognition system employing an analytical approach. The system is supported by a set of robust language independent features extracted on binary images. Parameters such as lower and upper baselines are used to derive a subset of baseline dependent features. Thus, word variability due to lower and upper parts of words is better taken into account. In addition, the proposed system learns character models without character pre-segmentation. Experiments that have been conducted on the benchmark IFN/ENIT database of Tunisian handwritten country/village names, show the advantage of the proposed approach and of the baseline-dependant features.

185 citations


Journal ArticleDOI
TL;DR: Using metafeatures, the scope of attribute-value learning is expanded to domains with instances that have some kind of recurring substructure, such as strokes in handwriting recognition, or local maxima in time series data.
Abstract: We present a method of constructive induction aimed at learning tasks involving multivariate time series data. Using metafeatures, the scope of attribute-value learning is expanded to domains with instances that have some kind of recurring substructure, such as strokes in handwriting recognition, or local maxima in time series data. The types of substructures are defined by the user, but are extracted automatically and are used to construct attributes. Metafeatures are applied to two real domains: sign language recognition and ECG classification. Using metafeatures we are able to generate classifiers that are either comprehensible or accurate, producing results that are comparable to hand-crafted preprocessing and comparable to human experts.

124 citations


Proceedings ArticleDOI
31 Aug 2005
TL;DR: An orientation independent technique for baseline detection of Arabic words is described and it is shown how the baseline can be exploited for slope and skew correction before proceeding with the steps of line and word separation.
Abstract: In order to improve the readability and the automatic recognition of handwritten document images, preprocessing steps are imperative. These steps in addition to conventional steps of noise removal and filtering include text normalization such as baseline correction, slant normalization and skew correction. These steps make the feature extraction process more reliable and effective. Recently Arabic handwriting recognition has received some attention from the research community. Due to the unique nature of the script, the conventional methods do not prove to be effective. In our work, we describe an orientation independent technique for baseline detection of Arabic words. In addition to that we describe, in the rest of the paper, our techniques for slant normalization, slope correction, line and word separation in handwritten Arabic documents. We show how the baseline can be exploited for slope and skew correction before proceeding with the steps of line and word separation.

91 citations


Patent
08 Dec 2005
TL;DR: In this paper, a handwriting recognition system is described, which consists of an input device having a handwriting input area and a recognition device configured to recognize the characters based on the strokes input by the input device, shapes of the strokes constructing two characters which are written successively and positional relations between or among the stroke constructing the two characters.
Abstract: A handwriting recognition apparatus is disclosed. In one embodiment the apparatus comprises an input device having a handwriting input area and configured to input a plurality of strokes constructing a plurality of characters written successively on the handwriting input area, and a recognition device configured to recognize the characters based on the strokes input by the input device, shapes of the strokes constructing two characters which are written successively and positional relations between or among the strokes constructing the two characters, whenever one stroke is input by the input device.

Journal ArticleDOI
TL;DR: A novel approach for the verification of the word hypotheses generated by a large vocabulary, offline handwritten word recognition system that has improved the word recognition rate as well as the reliability of the recognition system, while not causing significant delays in the recognition process.
Abstract: This paper presents a novel approach for the verification of the word hypotheses generated by a large vocabulary, offline handwritten word recognition system. Given a word image, the recognition system produces a ranked list of the N-best recognition hypotheses consisting of text transcripts, segmentation boundaries of the word hypotheses into characters, and recognition scores. The verification consists of an estimation of the probability of each segment representing a known class of character. Then, character probabilities are combined to produce word confidence scores which are further integrated with the recognition scores produced by the recognition system. The N-best recognition hypothesis list is reranked based on such composite scores. In the end, rejection rules are invoked to either accept the best recognition hypothesis of such a list or to reject the input word image. The use of the verification approach has improved the word recognition rate as well as the reliability of the recognition system, while not causing significant delays in the recognition process. Our approach is described in detail and the experimental results on a large database of unconstrained handwritten words extracted from postal envelopes are presented.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: A statistical method embedded with statistical features is proposed for Farsi/Arabic handwritten zip code recognition in this paper and thanks to statistically description of the skeleton, the nearest neighbor classifier can be used for recognition.
Abstract: A statistical method embedded with statistical features is proposed for Farsi/Arabic handwritten zip code recognition in this paper. The numeral is first smoothed and the skeleton is obtained. A set of feature points are then detected and the skeleton is decomposed into primitives. A primitive code includes the information of each primitive and a global code is derived from the primitive codes to describe the topological structure of the skeleton. By using the average and variance of X and Y changes in each primitive, the direction and curvature of the skeleton can be statistically described. Since the global codes have different lengths, we applied PCA algorithm to normalize their lengths. Thanks to statistically description of the skeleton, we can use the nearest neighbor classifier for recognition. According to experimental results, classification rate of 94.44% is obtained for numerals on the test sets gathered from various people with different educational background and different ages. Our database includes 480 samples per digit. We used 280 samples of each digit for training and the rest [200] for test.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: This paper deals with recognition of off-line unconstrained Oriya handwritten numerals and Neural network (NN) classifier and quadratic classifier are used separately for recognition and the results obtained from these two classifiers are compared.
Abstract: This paper deals with recognition of off-line unconstrained Oriya handwritten numerals. To take care of variability involved in the writing style of different individuals, the features are mainly considered from the contour of the numerals. At first, the bounding box of a numeral is segmented into few blocks and chain code histogram is computed in each of the blocks. Features are mainly based on the direction chain code histogram of the contour points of these blocks. Neural network (NN) classifier and quadratic classifier are used separately for recognition and the results obtained from these two classifiers are compared. We tested the result on 3850 data collected from different individuals of various background and we obtained 90.38% (94.81%) recognition accuracy from NN (quadratic) classifier with a rejection rate of about 1.84% (1.31%), respectively.

Journal ArticleDOI
TL;DR: A prototype system for automatic video-based whiteboard reading is presented and is designed for recognizing unconstrained handwritten text and is further characterized by an incremental processing strategy in order to facilitate recognizing portions of text as soon as they have been written on the board.
Abstract: The increasing popularity of whiteboards as a popular tool in meeting rooms has been accompanied by a growing interest in making use of the whiteboard as a user interface for human-computer interaction. Therefore, systems based on electronic whiteboards have been developed in order to serve as meeting assistants for, e.g., collaborative working. However, as special pens and erasers are required, natural interaction is restricted. In order to render this communication method more natural, it was proposed to retain ordinary whiteboard and pens and to visually observe the writing process using a video camera [22, 25]. In this paper a prototype system for automatic video-based whiteboard reading is presented. The system is designed for recognizing unconstrained handwritten text and is further characterized by an incremental processing strategy in order to facilitate recognizing portions of text as soon as they have been written on the board. We will present the methods employed for extracting text regions, preprocessing, feature extraction, and statistical modeling and recognition. Evaluation results on a writer-independent unconstrained handwriting recognition task demonstrate the feasibility of the proposed approach.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: This work proposes a novel algorithm that over-segments each word, and then removes extra breakpoints using knowledge of letter shapes, and annotates each detected letter with shape information, to be used for recognition in future work.
Abstract: We propose a novel algorithm for the segmentation and prerecognition of offline handwritten Arabic text. Our character segmentation method over-segments each word, and then removes extra breakpoints using knowledge of letter shapes. On a test set of 200 images, 92.3% of the segmentation points were detected correctly, with 5.1% instances of over-segmentation. The prerecognition component annotates each detected letter with shape information, to be used for recognition in future work.

Journal ArticleDOI
TL;DR: Technological advances have made possible new perspectives for signature recognition, by means of capturing devices which provide more than the simple signature image: pressure, acceleration, etc., making it even more difficult to forge a signature.
Abstract: A summarization of one of the most successful behavioral biometric recognition methods: signature recognition. Probably this is one of the oldest biometric recognition methods, with high legal acceptance. Technological advances have made possible new perspectives for signature recognition, by means of capturing devices which provide more than the simple signature image: pressure, acceleration, etc., making it even more difficult to forge a signature.

Journal ArticleDOI
TL;DR: Learning effects were found for using touch screen and handwriting recognition, and mouse and keyboard for browsing and searching tasks, and voice-menu assistance was associated with higher satisfaction for browsing tasks.
Abstract: This study investigated the effects of interaction devices on the Internet performance of novice older users, and ways to provide appropriate voice assistance to enhance browsing and searching performance of such users. Three experiments were designed and conducted to test three hypotheses. The results indicated that touch screen and handwriting recognition are better than mouse and keyboard in browsing time in the third trial. Touch screen was also found to be better in terms of performance time for keyword search tasks than mouse and voice input in the second trial, and is better in terms of user error for keyword search tasks than mouse and voice input in the first trial. Learning effects were found for using touch screen and handwriting recognition, and mouse and keyboard for browsing and searching tasks. Furthermore, voice-menu assistance was associated with higher satisfaction for browsing tasks.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: It is shown that one can achieve significantly better writer identification rates if one uses smaller feature subsets returned by different feature extraction and selection methods.
Abstract: To identify the author of a sample handwriting from a set of writers, 100 features are extracted from the handwriting sample. By applying feature selection and extraction methods on this set of features, subsets of lower dimensionality are obtained. We show that we can achieve significantly better writer identification rates if we use smaller feature subsets returned by different feature extraction and selection methods. The methods considered in this paper are feature set search algorithms, genetic algorithms, principal component analysis, and multiple discriminant analysis.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: A novel wavelet-based GGD method to replace the traditional 2D Gabor filters is presented, which not only achieves better experiment results but also greatly reduces the elapsed time on calculation.
Abstract: Handwriting-based writer identification is a hot research topic in the pattern recognition field. Nowadays, online handwriting-based writer identification is steadily growing toward its maturity. On the contrary, offline handwriting-based writer identification still remains as a challenging problem because writing features only can be extracted from the handwriting image in this situation. As a result, plenty of dynamic writing information, which is very valuable for writer identification, is lost. At present, 2D Gabor filter method is widely acknowledged as a good method for offline handwriting identification, however it still suffers from some inherent disadvantages, such as the high computational cost. In this paper, we present a novel wavelet-based GGD method to replace the traditional 2D Gabor filters. Shown in our experiments, this novel method not only achieves better experiment results but also greatly reduces the elapsed time on calculation.

01 Jan 2005
TL;DR: This paper discusses the use of Dynamic Time Warping for visually perceptive and intuitive character recognition and sets up a human factors experiment in which two variants of DTW are compared to a state-ofthe- art character classifier, HCLUS.
Abstract: This paper discusses the use of Dynamic Time Warping (DTW) for visually perceptive and intuitive character recognition. In particular in forensic document examination, techniques are required that yield results matching a human user's expectations. In our approach, the goal is to retrieve a set of best matching allographic prototypes based on a query input character. Since DTW compares each pair of closest points from two trajectories, our assumption is that it may resemble most of the pair-wise coordinate comparisons employed by humans. In order to assess our ideas, we have set up a human factors experiment in which two variants of DTW are compared to a state-ofthe- art character classifier, HCLUS. A number of 25 subjects judged the recognition results of these three classifiers for 130 queries. As a result, one particular implementation of DTW was significantly rated as the best system. Future research will combine these promising new findings with techniques that employ other distinctive features like those used by human experts. This research is embedded in the Dutch TRIGRAPH project, which pursues the design of forensic document examination techniques based on expert knowledge.

Proceedings ArticleDOI
19 Dec 2005
TL;DR: This paper focus on Urdu online handwriting recognition system that converts user hand written information into Urdu text based on analytical approach to segmentation for feature extraction, rule based slant analysis for slant removal and tree based dictionary search for classification.
Abstract: Handwriting has continued to persist as a mean of communication and recording information in day-to-day life even with the introduction of new technologies. Handwritten information can be used as computer input once it is converted to digital form. This paper focus on Urdu online handwriting recognition system that converts user hand written information into Urdu text. The working of recognition system is based on analytical approach to segmentation for feature extraction, rule based slant analysis for slant removal and tree based dictionary search for classification. The proposed tree search technique reduces the search space up to 96.2% and is therefore significantly faster than searching techniques, in which we have to process the whole dictionary to come up with the correct answer

01 Jan 2005
TL;DR: A new system for processing on-line whiteboard notes and using an off-line HMM-recognizer, which has been developed in the context of previous work, to achieve a statistically significant increase of the recognition rate.
Abstract: This paper introduces a new system for processing on-line whiteboard notes. Notes written on a whiteboard is a new modality in handwriting recognition research that has received relatively little attention in the past. For the recognition we use an off-line HMM-recognizer, which has been developed in the context of our previous work. The recognizer is supplemented with methods for processing the on-line data and generating the images. The system consists of six main modules: on-line preprocessing, transformation to off-line data, off-line preprocessing, feature extraction, classification and post-processing. The recognition rate of the basic recognizer in a writer independent experiment is 59,5%. By applying state-of-the-art methods, such as optimizing the number of states and Gaussian components, and by including a language model we could achieve a statistically significant increase of the recognition rate to 64.3%.

Journal ArticleDOI
01 Sep 2005
TL;DR: This paper presents an original two stages recognizer which is a model-based classifier which store an exhaustive set of character models and a pairwise classifiers which separate the most ambiguous pairs of classes.
Abstract: Handwriting recognition for hand-held devices like PDAs requires very accurate and adaptive classifiers. It is such a complex classification problem that it is quite usual now to make co-operate several classification methods. In this paper, we present an original two stages recognizer. The first stage is a model-based classifier which store an exhaustive set of character models. The second stage is a pairwise classifier which separate the most ambiguous pairs of classes. This hybrid architecture is based on the idea that the correct class almost systematically belongs to the two more relevant classes found by the first classifier. Experiments on a 80,000 examples database show a 30% improvement on a 62 classes recognition problem. Moreover, we show experimentally that such an architecture suits perfectly for incremental classification.

Proceedings ArticleDOI
06 Dec 2005
TL;DR: A novel approach to skew detection, correction as well as character segmentation has been presented for handwritten Bangla words as a test case and with the help of a candidate path one can handle both skew correction and segmentation successfully.
Abstract: Character segmentation is a necessary preprocessing step for character recognition in many handwritten word recognition systems. The most difficult case in character segmentation is the cursive script. Fully cursive nature of Bangla handwriting, the natural skewness in words poses some challenges for automatic character segmentation. In this article a novel approach to skew detection, correction as well as character segmentation has been presented for handwritten Bangla words as a test case. Segmenting points are extracted on the basis of some patterns observed in the handwritten words. With these segmenting points a graphical path (hereafter referred to as a candidate path) has been constructed. The handwritten words contain some consistent and also inconsistent skewness. Our algorithm can cope with both types of skewness at a time. Further the method is so direct that with the help of a candidate path one can handle both skew correction and segmentation successfully. the algorithm has been tested on a database prepared for laboratory use. The method yields fairly good results for this database.

Book ChapterDOI
01 Jan 2005
TL;DR: Key usability problems with the handwriting recognition interface are identified and classified, and some solutions are proposed in the form of design guidelines for both recognition-based and pen-based computer writing interfaces.
Abstract: This paper describes an empirical study with children that compared the three methods of writing — using pencil and paper, using the QWERTY keyboard at a computer, and using a pen and graphics tablet. The children wrote short stories. Where the graphics tablet was used, the text was recognized and presented to the children as ASCII text. Measures of user satisfaction, quantity of writing produced, and quality of writing produced were taken. In addition, the recognition process was evaluated by comparing what the child wrote with the resulting ASCII text. The results show that for the age group considered, writing at the tablet was as efficient as, and produced comparable writing to, the pencil and paper. The keyboard was less efficient. Key usability problems with the handwriting recognition interface are identified and classified, and we propose some solutions in the form of design guidelines for both recognition-based and pen-based computer writing interfaces.

Proceedings ArticleDOI
31 Aug 2005
TL;DR: This paper presents a point matching algorithm to learn the shape deformation characteristics of handwriting from real samples, and applies it to handwriting synthesis.
Abstract: Since it is extremely expensive to collect a large volume of handwriting samples, synthesized data are often used to enlarge the training set. We argue that, in order to generate good handwriting samples, a synthesis algorithm should learn the shape deformation characteristics of handwriting from real samples. In this paper, we present a point matching algorithm to learn the deformation, and apply it to handwriting synthesis. Preliminary experiments show the advantages of our approach.

Proceedings ArticleDOI
06 Dec 2005
TL;DR: A novel method for automatically segmenting and recognizing the various information fields present on a bank cheque and proposed four innovative features, namely; entropy, energy, aspect ratio and average fuzzy membership values.
Abstract: This paper describes a novel method for automatically segmenting and recognizing the various information fields present on a bank cheque. The uniqueness of our approach lies in the fact that it doesn’t necessitate any prior information and requires minimum human intervention. The extraction of segmented fields is accomplished by means of a connectivity based approach. For the recognition part, we have proposed four innovative features, namely; entropy, energy, aspect ratio and average fuzzy membership values. Though no particular feature is pertinent in itself but a combination of these is used for differentiating between the fields. Finally, a fuzzy neural network is trained to identify the desired fields. The system performance is quite promising on a large dataset of real and synthetic cheque images.

Journal ArticleDOI
TL;DR: It is shown experimentally that an offline recognition system using the recovered temporal order can achieve recognition performances that are much better than those obtained with the simple left-right order, and that come close to those of an online recognition system.

Patent
30 Dec 2005
TL;DR: In this article, a mixture of Bayesian networks (MBN) consisting of plural hypothesis-specific Bayesian Networks (HSBNs) having possibly hidden and observed variables was used for handwriting recognition.
Abstract: The invention performs handwriting recognition using mixtures of Bayesian networks. A mixture of Bayesian networks (MBN) consists of plural hypothesis-specific Bayesian networks (HSBNs) having possibly hidden and observed variables. A common external hidden variable is associated with the MBN, but is not included in any of the HSBNs. Each HSBN models the world under the hypothesis that the common external hidden variable is in a corresponding one of its states. The MBNs encode the probabilities of observing the sets of visual observations corresponding to a handwritten character. Each of the HSBNs encodes the probabilities of observing the sets of visual observations corresponding to a handwritten character and given a hidden common variable being in a particular state.