Showing papers on "Intelligent word recognition published in 2005"

ICDAR 2005 Arabic handwriting recognition 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.

IAM-OnDB - an on-line English sentence database acquired from handwritten text on a whiteboard

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.

Domain-based dialog speech recognition method and apparatus

Abstract: A domain-based speech recognition method and apparatus, the method including: performing speech recognition by using a first language model and generating a first recognition result including a plurality of first recognition sentences; selecting a plurality of candidate domains, by using a word included in each of the first recognition sentences and having a confidence score equal to or higher than a predetermined threshold, as a domain keyword; performing speech recognition with the first recognition result, by using an acoustic model specific to each of the candidate domains and a second language model and generating a plurality of second recognition sentences; and selecting at least one or more final recognition sentence from the first recognition sentences and the second recognition sentences. According to this method and apparatus, the effect of a domain extraction error by misrecognition of a word on selection of a final recognition result can be minimized.

Error correction for speech recognition systems

Abstract: Words in a sequence of words that is obtained from speech recognition of an input speech sequence are presented to a user, and at least one of the words in the sequence of words is replaced, in case it has been selected by a user for correction. Words with a low recognition confidence value are emphasized; alternative word candidates for the at least one selected word are ordered according to an ordering criterion; after replacing a word, an order of alternative word candidates for neighboring words in the sequence is updated; the replacement word is derived from a spoken representation of the at least one selected word by speech recognition with a limited vocabulary; and the word that replaces the at least one selected word is derived from a spoken and spelled representation of the at least one selected word.

Word recognition using ontologies

Abstract: Systems, and associated apparatus, methods, or computer program products, may use ontologies to provide improved word recognition. The ontologies may be applied in word recognition processes to resolve ambiguities in language elements (e.g., words) where the values of some of the characters in the language elements are uncertain. Implementations of the method may use an ontology to resolve ambiguities in an input string of characters, for example. In some implementations, the input string may be received from a language conversion source such as, for example, an optical character recognition (OCR) device that generates a string of characters in electronic form from visible character images, or a voice recognition (VR) device that generates a string of characters in electronic form from speech input. Some implementations may process the generated character strings by using an ontology in combination with syntactic and/or grammatical analysis engines to further improve word recognition accuracy.

An improved handwritten Chinese character recognition system using support vector machine

Abstract: This paper describes several techniques improving a Chinese character recognition system. Enhanced nonlinear normalization, feature extraction and tuning kernel parameters of support vector machine on a large data set with thousands of classes, contribute to improvement of the overall system performance. The enhanced nonlinear normalization method not only solves the aliasing problem in the original Yamada et al.'s nonlinear normalization method but also avoids the undue stroke distortion in the peripheral region of the normalized image. The support vector machine is for the first time tested on a large data set composed of several million samples and thousands of classes. The recognition system has achieved a high recognition rate of 99.0% on ETL9B, a handwritten Chinese character database.

Pre-processing methods for handwritten Arabic documents

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.

Machine recognition of online handwritten Devanagari characters

Abstract: In this paper, we describe a system for the automatic recognition of isolated handwritten Devanagari characters obtained by linearizing consonant conjuncts. Owing to the large number of characters and resulting demands on data acquisition, we use structural recognition techniques to reduce some characters to others. The residual characters are then classified using the subspace method. Finally the results of structural recognition and feature-based matching are mapped to give final output. The proposed system is evaluated for the writer dependent scenario.

Recognition and verification of unconstrained handwritten words

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.

Oriya handwritten numeral recognition system

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.

Boosted decision trees for word recognition in handwritten document retrieval

Abstract: Recognition and retrieval of historical handwritten material is an unsolved problem. We propose a novel approach to recognizing and retrieving handwritten manuscripts, based upon word image classification as a key step. Decision trees with normalized pixels as features form the basis of a highly accurate AdaBoost classifier, trained on a corpus of word images that have been resized and sampled at a pyramid of resolutions. To stem problems from the highly skewed distribution of class frequencies, word classes with very few training samples are augmented with stochastically altered versions of the originals. This increases recognition performance substantially. On a standard corpus of 20 pages of handwritten material from the George Washington collection the recognition performance shows a substantial improvement in performance over previous published results (75% vs 65%). Following word recognition, retrieval is done using a language model over the recognized words. Retrieval performance also shows substantially improved results over previously published results on this database. Recognition/retrieval results on a more challenging database of 100 pages from the George Washington collection are also presented.

Segmentation and pre-recognition of Arabic handwriting

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.

Camera-based Kanji OCR for mobile-phones: practical issues

Abstract: A camera based optical character reader (OCR) for Japanese Kanji characters was implemented on a mobile phone. This OCR has three key features. The first is discriminative feature extraction (DFE) which enables a character classifier needing only small memory size. The second is a word segmentation method specially designed for looking up Japanese words in a dictionary. The third feature is a GUI suitable for a mobile phone. A prototype mobile phone Kanji OCR was constructed and experimentally tested. Recognition accuracy of over 95% was obtained under the best conditions, which shows the potential of our prototype as a new type of electronic dictionary.

Arabic Handwritten Words Recognition Based on a Planar Hidden Markov Model.

Abstract: Off-line recognition of handwritten words is a difficult task due to the high variability and uncertainty of human writing. The majority of the recent systems are constrained by the size of the lexicon to deal with and the number of writers. In this paper, we propose an approach for multi-writers Arabic handwritten words recognition. The developed method uses multiple sources of information at the description and the classification levels. A hybrid planar Markovien modelling permitting to follow the horizontal and vertical variations of the writing has been adopted. This modelling is based on different levels of segmentation: horizontal, natural and vertical. The process of segmentation conducts to the decomposition of the writing in a limited set of elementary entities, with simplified morphologies specific to every horizontal band. The choice of different type of primitives is then imposed in order to assure an efficient description. Different architectures of modelling proved also to be indispensable. The classification is finally achieved using a Planar Hidden Markov Model.

A Model of On-line Handwritten Japanese Text Recognition Free from Line Direction and Writing Format Constraints

Abstract: This paper presents a model and its effect for on-line handwritten Japanese text recognition free from line-direction constraint and writing format constraint such as character writing boxes or ruled lines. The model evaluates the likelihood composed of character segmentation, character recognition, character pattern structure and context. The likelihood of character pattern structure considers the plausible height, width and inner gaps within a character pattern that appear in Chinese characters composed of multiple radicals (subpatterns). The recognition system incorporating this model separates freely written text into text line elements, estimates the average character size of each element, hypothetically segments it into characters using geometric features, applies character recognition to segmented patterns and employs the model to search the text interpretation that maximizes likelihood as Japanese text. We show the effectiveness of the model through recognition experiments and clarify how the newly modeled factors in the likelihood affect the overall recognition rate.

Speech recognition assisted autocompletion of composite characters

Abstract: Speech recognition assisted autocompletion of textual composite words or characters (i.e. words or characters containing a number of components) is provided. In response to user input specifying a component of a word or character, a list of candidate words or characters is generated. The desired word or character can be selected, or the list of candidate words or characters can be narrowed, in response to the user speaking the desired word or character. As a result, entry of words or characters formed from a number of letters, strokes, or word shapes is facilitated by user input comprising a combination of a specification of a component of the desired word or character and speech corresponding to a pronunciation of the desired word or character.

Tree structure for word extraction from handwritten text lines

Abstract: Word extraction from handwritten text lines usually involves the calculation of a line specific threshold which separates the gaps between words from the gaps inside the words in that line. We show that this approach can be improved if the decision about a gap is not only made in terms of a threshold, but also depends on the context of that gap, i.e. if the relative sizes of the surrounding gaps are taken into consideration. For this purpose, we propose to build a structure tree of the text line, whose nodes represent possible word candidates. Such a tree is traversed in a top-down manner to find the nodes that correspond to words of the text line. Experiments with different gap metrics as well as threshold types show that the new method can yield significant improvements over conventional word extraction methods.

Preprocessing and Image Enhancement Algorithms for a Form-based Intelligent Character Recognition System

Abstract: A Form-based Intelligent Character Recognition (ICR) System for handwritten forms, besides others, includes functional components for form registration, character image extraction and character image classification. Needless to say, the classifier is a very important component of the ICR system. Automatic recognition and classification of handwritten character images is a complex task. Neural Networks based classifiers are now available. These are fairly accurate and demonstrate a significant degree of generalisation. However any such classifier is highly sensitive to the quality of the character images given as input. Therefore it is essential that the preprocessing components of the system, form registration and character image extraction, are well designed. In this paper we discuss the form image registration technique and the image masking and image improvement techniques implemented in our system as part of the character image extraction process. These simple yet effective techniques help in preparing the input character image for the neural networks-based classifiers and go a long way in improving overall system accuracy. Although these algorithms have been discussed with reference to our ICR system they are generic in their applicability and may find use in other scenarios as well.

A Front-End OCR for Omni-Font Persian/Arabic Cursive Printed Documents

Abstract: Compared to non-cursive scripts, optical character recognition of cursive documents comprises extra challenges in layout analysis as well as recognition of the printed scripts. This paper presents a front-end OCR for Persian/Arabic cursive documents, which utilizes an adaptive layout analysis system in addition to a combined MLP-SVM recognition process. The implementation results on a comprehensive database show a high degree of accuracy which meets the requirements of commercial use.

Voice Recognition Device and Method, and Program

Abstract: A speech recognition system in which a user may correct a recognition error resulting from speech recognition more efficiently and easily. Speech recognition means compares a plurality of words inputted from speech input means with a plurality of words stored in dictionary means, respectively, and determines a most-competitive word candidate. Word correction means has a word correction function of correcting the words constituting a word sequence displayed on a screen. Competitive word display commanding means selects one or more competitive words having competitive probabilities close to the competitive probability of the most-competitive word candidate and displays the one or more competitive words adjacent to the most-competitive word candidate. Competitive word selection means selects an appropriate correction word from the one or more competitive words. Word replacement commanding means causes one of the most-competitive word candidate to be replaced with the correction word selected by the competitive word selection means.

A Novel Approach to Skew Detection and Character Segmentation for Handwritten Bangla Words

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.

Affixal approach for Arabic decomposable vocabulary recognition a validation on printed word in only one font

Abstract: We propose a new approach for Arabic word recognition called affixal approach. This approach is founded on morphological structure of Arabic vocabulary. A mechanism of decomposition-recognition is used in our approach and makes it possible to lead to a set of reliable solutions for each word. This mechanism tries to recognize word basic morphemes: prefix, infix, suffix and root contrary to existing approaches which are usually based on recognition of word entity by holistic approach, pseudo-word entity by pseudo-analytical approach or letter entity by analytical approach. In this paper, we will present limits of existing approaches for Arabic word recognition. We will expose then Arabic vocabulary structure. We will detail after affixal approach for Arabic decomposable vocabulary recognition with a word example. Lastly, we will expose experimental results obtained on a basis of 1000 words data set.

Dynamic time warping applied to Tamil character recognition

Abstract: This paper describes the use of dynamic time warping (DTW) for classifying handwritten Tamil characters. Since DTW can match characters of arbitrary length, it is particularly suited for this domain. We built a prototype based classifier that uses DTW both for generating prototypes and for calculating a list of nearest prototypes. Prototypes were automatically generated and selected. Two tests were performed to measure the performance of our classifier in a writer dependent, and in a writer independent setting. Furthermore, several strategies were developed for rejecting uncertain cases. Two different rejection variables were implemented and using a Monte Carlo simulation, the performance of the system was tested in various configurations. The results are promising and show that the classifier can be of use in both writer dependent and writer independent automatic recognition of handwritten Tamil characters.

Unconstrained handwritten character recognition using metaclasses of characters

Abstract: In this paper we tackle the problem of unconstrained handwritten character recognition using different classification strategies. For such an aim, four multilayer perceptron classifiers (MLP) were built and used into three different classification strategies: combination of two 26-class classifiers; 26-metaclass classifier; 52-class classifier. Experimental results on the NIST SD19 database have shown that the recognition rate achieved by the metaclass classifier (87.8%) outperforms the other approaches (82.9% and 86.3%).

Hybrid OCR combination approach complemented by a specialized ICR applied on ancient documents

Abstract: In spite of the improvement of commercial optical character recognition (OCR) during the last years, their ability to process different kinds of documents can also be a default. They cannot produce a perfect recognition for all documents. However they allow producing high result for standard cases. We propose in this paper a model combining several OCRs and a specialized ICR (intelligent character recognition) based on a convolutional neural network to complement them. Instead of just performing several OCRs in parallel and applying a fusing rule of the results, a specialized neural network with an adaptive topology is added to complement the OCRs in function of the OCRs errors. This system has been tested on ancient documents containing old characters and old fonts not used in contemporary documents. The OCRs combination increases the recognition of about 3% whereas the ICR improves the recognition of rejected characters of more than 5%.

Off-line handwritten character recognition by SVM based on the virtual examples synthesized from on-line characters

Abstract: This paper proposes a method to improve the off-line character classifiers which are learned from examples by using the virtual examples synthesized from on-line character database. To obtain the good classifiers, usually a large database which contains enough number of variations of handwritten characters is required. However, in practice collecting enough number of data is time-consuming and costly. In this paper, we propose a method to train SVM for off-line character recognition based on the artificially augmented examples using on-line characters. In our method, a virtual example is synthesized from an on-line character by applying affine transformation to each stroke. SVM classifiers are trained by using the artificially generated patterns. To reject inappropriate artificial examples, a preliminary SVM is learned from the original set of samples, and then used for data selection. Using the augmented training samples, the final SVM is obtained. We examine the effectiveness of the proposed method by experiments of handwritten Japanese Hiragana character classification.

A human interactive proof algorithm using handwriting recognition

Abstract: The recognition of unconstrained handwriting continues to be a difficult task for computers despite active research for several decades. This is because handwritten text offers great challenges such as: character and word segmentation, character recognition, variation between handwriting styles, different character size and orientation, no font constraints, the type of printing surface, as well as the background clarity. In this paper, we explore the gap in the ability in reading handwritten text between humans and computers to propose solutions for security problems in Web services. We present a new HIP algorithm that uses handwriting recognition task to distinguish between humans and computers. We propose methods to deform handwritten text images to make them indecipherable by computers and explore the cognitive factors that assist humans in reading and understanding. Experimental results on both humans and computers are presented and compared.

Statistical language models for on-line handwritten sentence recognition

Abstract: This paper investigates the integration of a statistical language model into an on-line recognition system in order to improve word recognition in the context of handwritten sentences. Two kinds of models have been considered: n-gram and n-class models (with a statistical approach to create word classes). All these models are trained over the Susanne corpus and experiments are carried out on sentences from this corpus which were written by several writers. The use of a statistical language model is shown to improve the word recognition rate and the relative impact of the different language models is compared. Furthermore, we illustrate the interest to define an optimal cooperation between the language model and the recognition system to re-enforce the accuracy of the system.

Recognition for large sets of handwritten mathematical symbols

Abstract: Natural and convenient mathematical handwriting recognition requires recognizers for large sets of handwritten symbols. This paper presents a recognition system for such handwritten mathematical symbols. We use a pre-classification strategy, in combination with elastic matching, to improve recognition speed. Elastic matching is a model-based method that involves computation proportional to the set of candidate models. To solve this problem, we prune prototypes by examining character features. To this end, we have defined and analyzed different features. By applying these features into an elastic recognition system, the recognition speed is improved while maintaining high recognition accuracy.

Apparatus and method for handwriting recognition

Abstract: A method and device is provided for recognizing characters in a handwritten input representing an input character string. A character sub-string preceding an unrecognized character in the input character string is determined. Handwriting recognition is used to provide one or more candidate characters for the unrecognized character. One of the one or more candidate characters is then selected. The candidate character selected, is the one which is most likely to be a correct recognition of the unrecognized character based on the determined character sub-string.