Online Handwritten Character Recognition Research Articles

An irrelevant variability normalization approach to discriminative training of multi-prototype based classifiers and its applications for online handwritten Chinese character recognition

This paper presents an irrelevant variability normalization (IVN) approach to jointly discriminative training of feature transforms and multi-prototype based classifier for recognition of online handwritten Chinese characters. A sample separation margin based minimum classification error criterion is adopted in IVN-based training, while an Rprop algorithm is used for optimizing the objective function. For the IVN approach based on piecewise linear transforms, the corresponding recognizer can be made both compact and efficient by using a two-level fast-match tree whose internal nodes coincide with the labels of feature transforms. Furthermore, the IVN system using weighted sum of linear transforms outperforms that based on piecewise linear transforms. The effectiveness of the proposed approach is first confirmed using an in-house developed online Chinese handwriting corpus with a vocabulary of 9306 characters, and then further verified on a standard benchmark database for an online handwritten character recognition task with a vocabulary of 3755 characters.

Handwritten Character Recognition in Malayalam Scripts – A Review

Handwritten character recognition is one of the most challenging and ongoing areas of research in the field of pattern recognition. HCR research is matured for foreign languages like Chinese and Japanese but the problem is much more complex for Indian languages. The problem becomes even more complicated for South Indian languages due to its large character set and the presence of vowels modifiers and compound characters. This paper provides an overview of important contributions and advances in offline as well as online handwritten character recognition of Malayalam scripts.

Finding Critical Points of Handwritten Persian/Arabic Character

On-line handwritten character recognition system is a special line of research in image processing and pattern recognition field, it can also considered a special process of in academic researches and production fields in the past decade. The general steps in a recognition system are pre-processing, segmentation, feature extraction and classifications. The techniques of pre-processing stage play an excessive role in the system and directly affect the system performance. The focus of this article is on doing pre-processing stage and providing a most desirable data set form the raw data using in feature extraction and then to increase the rate of recognition system. For this reason, the novel algorithm is presented for finding a critical points set of hand-drawn letters. These critical points help extracting the correct structural and statistical features of on-line character handwriting in any given style writing.

Efficient online handwritten Chinese character recognition system using a two-dimensional functional relationship model

Efficient online handwritten Chinese character recognition system using a two-dimensional functional relationship modelThis paper presents novel feature extraction and classification methods for online handwritten Chinese character recognition (HCCR). TheX-graph andY-graph transformation is proposed for deriving a feature, which shows useful properties such as invariance to different writing styles. Central to the proposed method is the idea of capturing the geometrical and topological information from the trajectory of the handwritten character using theX-graph and theY-graph. For feature size reduction, the Haar wavelet transformation was applied on the graphs. For classification, the coefficient of determination (R2p) from the two-dimensional unreplicated linear functional relationship model is proposed as a similarity measure. The proposed methods show strong discrimination power when handling problems related to size, position and slant variation, stroke shape deformation, close resemblance of characters, and non-normalization. The proposed recognition system is applied to a database with 3000 frequently used Chinese characters, yielding a high recognition rate of 97.4% with reduced processing time of 75.31%, 73.05%, 58.27% and 40.69% when compared with recognition systems using the city block distance with deviation (CBDD), the minimum distance (MD), the compound Mahalanobis function (CMF) and the modified quadratic discriminant function (MQDF), respectively. High precision rates were also achieved.

HMM-based online handwritten gurmukhi character recognition

This paper presents a hidden Markov model-based online handwritten character recognition for Gurmukhi script. We discuss a procedure to develop a hidden Markov model database in order to recognize Gurmukhi characters. A test with 60 handwritten samples, where each sample includes 41 Gurmukhi characters, shows a 91.95% recognition rate, and an average recognition speed of 0.112 seconds per stroke. The hidden Markov model database has been developed in XML using 5330 Gurmukhi characters. This work shall be useful to implement a hidden Markov model in online handwriting recognition and its software development.

Modeling inverse covariance matrices by expansion of tied basis matrices for online handwritten Chinese character recognition

The state-of-the-art modified quadratic discriminant function (MQDF) based approach for online handwritten Chinese character recognition (HCCR) assumes that the feature vectors of each character class can be modeled by a Gaussian distribution with a mean vector and a full covariance matrix. In order to achieve a high recognition accuracy, enough number of leading eigenvectors of the covariance matrix have to be retained in MQDF. This paper presents a new approach to modeling each inverse covariance matrix by basis expansion, where expansion coefficients are character-dependent while a common set of basis matrices are shared by all the character classes. Consequently, our approach can achieve a much better accuracy–memory tradeoff. The usefulness of the proposed approach to designing compact HCCR systems has been confirmed and demonstrated by comparative experiments on popular Nakayosi and Kuchibue Japanese character databases.

A large margin approach for writer independent online handwriting classification

This paper proposes a new approach for classifying multivariate time-series with applications to the problem of writer independent online handwritten character recognition. Each time-series is approximated by a sum of piecewise polynomials in a suitably defined Reproducing Kernel Hilbert Space (RKHS). Using the associated kernel function a large margin classification formulation is proposed which can discriminate between two such functions belonging to the RKHS. The associated problem turns out to be an instance of convex quadratic programming. The resultant classification scheme applies to many time-series discrimination tasks and shows encouraging results when applied to online handwriting recognition tasks.

Prototype learning for structured pattern representation applied to on-line recognition of handwritten Japanese characters

This paper describes prototype learning for structured pattern representation with common subpatterns shared among multiple character prototypes for on-line recognition of handwritten Japanese characters. Prototype learning algorithms have not yet been shown to be useful for structured or hierarchical pattern representation. In this paper, we incorporate cost-free parallel translation to negate the location distributions of subpatterns when they are embedded in character patterns. Moreover, we introduce normalization into a prototype learning algorithm to extract true feature distributions in raw patterns to aggregate distributions of feature points to subpattern prototypes. We show that our proposed method significantly improves structured pattern representation for Japanese on-line character patterns.

Adaptive combination of adaptive classifiers for handwritten character recognition

In this paper we examine the feasibility of combining two distinct layers of on-line adaptation for improving overall handwritten character recognition performance. These two approaches are adaptive classifiers and an adaptive committee used to combine them. On-line adaptive handwritten character classifiers are first discussed and the significant performance enhancements they can provide illustrated. We then examine the benefits from combining classifiers for this task, adaptive and non-adaptive, and present an adaptive committee structure suitable for this doubly adaptive framework. Experiments in combining the two adaptation approaches to form an adaptive committee consisting of adaptive member classifiers are described. The results show that while adaptation of the individual classifiers provides on average the most benefit in comparison to the non-adaptive reference level, the use of an adaptive combination of adaptive classifiers is still capable of enhancing the recognition performance by a significant margin. The usefulness of the proposed doubly adaptive approach is in this paper demonstrated in the domain of on-line handwritten character recognition, but we argue that the proposed methodology could also be applied to other application domains.

Writer Independent Online Handwritten Character Recognition Using a Simple Approach

This study describes the simple approach involved in online handwriting recognition. Conventionally, the data obtained needs a lot of preprocessing including filtering, smoothing, slant removing and size normalization before recognition process. Instead of doing such lengthy preprocessing, this study presents a simple approach to extract the useful character information. The whole process requires no preprocessing and size normalization. This research evaluates the use of the Back-propagation Neural Network (BPN) and presents feature extraction mechanism in full detail to work with on-line handwriting recognition. The obtained recognition rates were 51 to 83% using the BPN for different sets of character samples. This study also describes a performance study in which a recognition mechanism with multiple thresholds is evaluated for back-propagation architecture. The results indicate that the application of multiple thresholds has significant effect on recognition mechanism. This is a writer-independent system and the method is applicable for off-line character recognition as well. The technique is tested for upper-case English alphabets for a number of different styles from different subjects.

Automatic fuzzy rule base generation for on-line handwritten alphanumeric character recognition

The main challenge in handwritten character recognition involves the development of a method that can generate descriptions of the handwritten objects in a short period of time. Due to its low computational requirement, fuzzy logic is probably the most efficient method available for on-line character recognition. The most tedious task associated with using fuzzy logic for online character recognition involves the building of the rule-base that would describe the characters to be recognized. This problem is further complicated by the fact that different people write the same character in completely different ways. This paper describes a method that can be used to automatically generate a database containing the fuzzy membership values that describe the characters written by different individuals.

Special Issue: Selected papers from the KES2004 conference

The Eighth International Conference on Knowledge Based Intelligent Information and Engineering Systems was held at the Intercontinental Hotel, Wellington, hosted by Wellington Institute of Technology, New Zealand in September 2004. KES2004 aimed to provide a high-tech forum for the presentation of recent research into the theory and applications of intelligent systems and techniques. However, it also focused on some significant emerging intelligent technologies including evolvable hardware, evolutionary computation in computational intelligence, DNA computing, artificial immune systems, bioinformatics using intelligent and machine learning techniques, and intelligent web mining. The conference attracted about 500 delegates from 55 countries and the proceedings contained approximately 500 papers. This Special Issue contains extended versions of nine papers presented at KES2004, selected for qualities of innovation, application of leading-edge intelligent techniques, or overall excellent quality research. The first paper,by A. Kusiak, A. Burns and F. Milster, describes a data-mining approach applied to the analysis of parameters relating to a circulating fluidisedbed boiler. The outcome of the research has interesting implications on the direction of research into the optimisation of energy production. The second paper, by B. Kostek and J. Wojcik, describes innovative work in which techniques often used in data-mining were applied to improve the effectiveness of the retrieval of stored musical rhythms. The next paper, by J.A. Rose, describes work relating to recent developments in DNA computing. Theory and results are presented. The fourth paper, by M.F. Ursu, B. Virginas, G. Owusu and C. Voudouris, describes an approach to workforce allocation, modelled as a distributed system. The work combines an agent-based model combined with rulebased expressions in an original combination. Good global solutions are obtained from the distributed algorithm. The fifth paper, by V.K. Murthy, describes research in which contextual knowledge management in peer to peer computing is applied to mobile-multiplayer games and robotics. Paper number six, by M. Ong, X. Ren, G. Allan, V. Kadirkamanathan, H.A. Thompson and P.J. Fleming, presents a practical framework under which to build a decision support system using a Grid computing paradigm. The system is applied to aeroengine monitoring. The next paper, authored by R. Ranawa, V. Palade and G.E.M.D.C. Bandara, describes an approach to the automatic generation of a fuzzy rule base for on-line hand-written alphanumeric character recognition. The method was found by the authors to be reliable and simple. The penultimate paper of the selection, written by D. Kim, N.-H. Kim, S.-J. Seo and G.-T. Park, describes simulation-based work that uses a fuzzy system to effectively model a practical walking bipedal robot. The final paper of the special issue, authored by V. Gorodetsky, O. Karsaev and V. Samoilov, describes an intelligent systems based generic approach to the on-line updating of situation assessment. We would like to thank the authors for informing us of the results of their work through their papers. We would also like to thank the reviewers for their comments, which resulted in improvements in the papers. We hope that readers appreciate from the papers some of the challenges of modern intelligent systems research, and some of the approaches that are being adopted to overcome them.

Development and evaluation of a hierarchical clustering and pattern averaging method applicable to on‐line handwritten character recognition based on feature point approximation

AbstractWith the approximate matching methods based on feature points that have become widely used in on‐line character recognition in recent years, it is difficult to collect statistics for individual points as the number of points used to match a single character varies from pattern to pattern. It has therefore not been possible to apply simultaneous clustering and pattern averaging techniques that have proven effective elsewhere in improving recognition rates while controlling the size of a recognition dictionary. In this paper, to deal with this situation, we note that it is possible to use the matching function of a recognition engine to find the average value between two patterns. We propose a method based on hierarchical clustering that allows the simultaneous clustering and averaging of patterns to create a dictionary and is also applicable to on‐line character recognition using feature point approximation methods. We have conducted experiments using the handwriting database of Tokyo University of Agriculture and Technology. In these experiments using samples of handwriting from 20 people, we found that by optimizing a cluster merge condition threshold parameter Lth and a recording condition threshold parameter N we were able to obtain good recognition rates: 88.1% for non‐kanji characters, 97.6% for 500 kanji characters with multiple dictionary entries, and 91.1% for 500 single entry kanji characters. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 88(12): 58–68, 2005; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20209

Methods for adaptive combination of classifiers with application to recognition of handwritten characters

This paper discusses two techniques for improving the recognition accuracy for online handwritten character recognition: committee classification and adaptation to the user. Combining classifiers is a common method for improving recognition performance. Improvements are possible because the member classifiers may make different errors. Much variation exists in handwritten characters, and adaptation is one feasible way of dealing with such variation. Even though adaptation is usually performed for single classifiers, it is also possible to use adaptive committees. Some novel adaptive committee structures, namely, the dynamically expanding context (DEC), modified current best learning (MCBL), and class-confidence critic combination (CCCC), are presented and evaluated. They are shown to be able to improve on their member classifiers, with CCCC offering the best performance. Also, the effect of having either more or less diverse sets of member classifiers is considered.

A new feature extraction technique for on-line recognition of handwritten alphanumeric characters

In this work a new method of feature extraction for an interactive and adaptive recognizer for on-line handwritten alphanumeric characters has been proposed. The system is suitable for use in conjunction with magnetic pen based devices for inputting data to a data processing system or a computer terminal. The features are extracted from dynamically changing locations of the writing device. The new method of feature extraction is simple, computationally light and fast enough for adaptive on-line use. Extracted features are robust with respect to all possible distortions like shape, size, and orientation. For simulation experiment, numerals 0–9 are used. A single hidden layer feed forward neural network trained by Quickprop algorithm, a variation of error back propagation is used for recognition. Very high recognition rates, even for highly distorted samples have been achieved confirming high generalization capability of the extracted feature set.

필기체 한글 문자 인식을 위한 획 추출에 관한 연구

필기체 문자 인식은 온라인 필기체 문자 인식과 오프라인 필기체 문자 인식으로 나누어진다. 온라인 필기체 문자 인식은 타블렛과 같은 펜 기반의 전자식 입력 장치를 이용하여 필기의 순서와 획의 위치와 같은 동적인 필기 정보를 문자의 입력 시 획득할 수 있어 오프라인 필기체 문자 인식에 비해 큰 연구 성과를 이루었다. 그러나 오프라인 필기체 문자 인식은 온라인 필기체 문자 인식에서와 같이 동적인 정보를 입력받을 수 없고, 다양한 필기와 자소의 겹침이 심하며 획 사이의 잡영을 많이 가지고 있어 인식의 전처리 결과에 따라 인식 성능이 크게 달라진다. 본 논문에서는 오프라인 필기체 한글 문자 인식을 위해 문자의 동적인 정보를 포함하는 획을 효과적으로 추출하는 방법을 제안한다. 제안된 방법은 전처리 과정으로 먼저 Watershed 알고리즘을 이용하여 입력된 필기체 문자 영상의 향상 및 이진화를 수행한다. 이진화된 문자부를 변형된 Lu와 Wang의 세선화 알고리즘을 사용하여 세선화를 수행한 후 문자에서의 특징점을 추출하여 세그먼트 화소열을 추출하고, 최대 허용 오차법을 이용하여 벡터화한다. 벡터화의 수행으로 몇 개의 획이 하나의 세그먼트로 묶인 경우, 하나의 세그먼트 화소열은 2 또는 그 이상의 세그먼트 벡터로 분리된다. 추출된 세그먼트 벡터들을 완전한 획으로 재구성하기 위해서 오른손 필기 좌표계 시스템을 이용하여 벡터의 방향적인 성분을 인간의 필기 획의 방향에 알맞게 수정하고, 수정된 세그먼트 벡터의 방향성과 분기 정보를 이용하여 인접한 결합 가능한 세그먼트 벡터를 결합함으로써 문자 인식에 적합한 완전한 획으로 재구성한다. 실험 결과 제안된 방법이 필기체 한글 문자 인식에 적합함을 알 수 있었다. Handwritten character recognition is classified into on-line handwritten character recognition and off-line handwritten character recognition. On-line handwritten character recognition has made a remarkable outcome compared to off-line hacdwritten character recognition. This method can acquire the dynamic written information such as the writing order and the position of a stroke by means of pen-based electronic input device such as a tablet board. On the contrary, Any dynamic information can not be acquired in off-line handwritten character recognition since there are extreme overlapping between consonants and vowels, and heavily noisy images between strokes, which change the recognition performance with the result of the preprocessing. This paper proposes a method that effectively extracts the stroke including dynamic information of characters for off-line Korean handwritten character recognition. First of all, this method makes improvement and binarization of input handwritten character image as preprocessing procedure using watershed algorithm. The next procedure is extraction of skeleton by using the transformed Lu and Wang's thinning: algorithm, and segment pixel array is extracted by abstracting the feature point of the characters. Then, the vectorization is executed with a maximum permission error method. In the case that a few strokes are bound in a segment, a segment pixel array is divided with two or more segment vectors. In order to reconstruct the extracted segment vector with a complete stroke, the directional component of the vector is mortified by using right-hand writing coordinate system. With combination of segment vectors which are adjacent and can be combined, the reconstruction of complete stroke is made out which is suitable for character recognition. As experimentation, it is verified that the proposed method is suitable for handwritten Korean character recognition.

Development of an online handwritten character recognition module on WWW client and its implementation in a WWW‐based intelligent tutoring system

AbstractInterfaces for handwritten inputs are expected to be useful as means by which a student can use natural handwritten inputs in an intelligent tutoring system. Intelligent tutoring systems operating on the WWW have been developed with the input interfaces of the mouse and keyboard. The authors have developed HandChaRM‐WWW, which is a module performing handwritten character recognition on the WWW. The module is written in Java and installed on a WWW client such that no communication delay occurs during character recognition. The characters recognized are limited to numbers, operators, the equal sign, and the like, and handwritten character recognition is realized by extracting characters with high speed and high accuracy. WWW FRONTIER has been developed by implementing HandChaRM‐WWW on an intelligent CAI system which guides fractional calculations. WWW FRONTIER is a drill‐type intelligent tutoring system which responds to a student by giving a fractional adding or subtracting problem. A student can input a fractional equation on the WWW in handwritten form. A handwritten character recognition rate of 93% has been obtained in evaluation experiments, which constitutes a practical level. The architecture of HandChaRM‐WWW is considered to be effective in intelligent tutoring systems for other domains which have characters as targets of recognition. © 2002 Scripta Technica, Syst Comp Jpn, 33(2): 12–21, 2002

A novel invariant mapping applied to hand-written arabic character recognition

This paper describes an application of a novel mapping, one that is intended for use in on-line hand-written character recognition. This mapping produces the same output pattern regardless of the orientation, position, and size of the input pattern. The mapping has the advantage of being simple. This makes it computationally efficient and fast, which in turn makes it appropriate for on-line implementations. To demonstrate the usefulness of this mapping, a recognition system utilizing it has been developed for hand-written Arabic characters. The performance of this system is shown to be comparable to that of the existing on-line Arabic character recognition systems.

A Set of Neural Tools for Human-Computer Interactions: Application to the Handwritten Character Recognition, and Visual Speech Recognition Problems

This paper presents a new technique of data coding and an associated set of homogenous processing tools for the development of Human Computer Interactions (HCI). The proposed technique facilitates the fusion of different sensorial modalities and simplifies the implementations. The coding takes into account the spatio-temporal nature of the signals to be processed in the framework of a sparse representation of data. Neural networks adapted to such a representation of data are proposed to perform the recognition tasks. Their development is illustrated by two examples: one of on-line handwritten character recognition; and the other of visual speech recognition.

A new linguistic decoding method for online handwritten Chinese character recognition

This paper presents a new linguistic decoding method for online handwritten Chinese character recognition. The method employs a hybrid language model which combinesN-gram and linguistic rules by rule quantification technique. The linguistic decoding algorithm consists of three stages: word lattice construction, the optimal sentence hypothesis search and self-adaptive learning mechanism. The technique has been applied to palmtop computer’s online handwritten Chinese character recognition. Samples containing millions of characters were used to test the linguistic decoder. In the open experiment, accuracy rate up to 92% is achieved, and the error rate is reduced by 68%.