Class-specific feature polynomial classifier for pattern classification and its application to handwritten numeral recognition

The polynomial classifier (PC) that takes the binomial terms of reduced subspace features as inputs has shown superior performance to multilayer neural networks in pattern classification. In this paper, we propose a class-specific feature polynomial classifier (CFPC) that extracts class-specific features from class-specific subspaces, unlike the ordinary PC that uses a class-independent subspace. The CFPC can be viewed as a hybrid of ordinary PC and projection distance method. The class-specific features better separate one class from the others, and the incorporation of class-specific projection distance further improves the separability. The connecting weights of CFPC are efficiently learned class-by-class to minimize the mean square error on training samples. To justify the promise of CFPC, we have conducted experiments of handwritten digit recognition and numeral string recognition on the NIST Special Database 19 (SD19). The digit recognition task was also benchmarked on two standard databases USPS and MNIST. The results show that the performance of CFPC is superior to that of ordinary PC, and is competitive with support vector classifiers (SVCs).     

手写混合字符集识别的多特征多级分类器设计

针对常用的银行汉字和阿拉伯数字混合字符集的识别,提出了依据不同的分类要求，分别选取不同的分类特征，并采用先聚类再用多层感知器(MLP)神经网络分类的多级分类器进行识别的设计方法。实验结果表明，该方法用于手写体混合字符集的识别是行之有效的。     

An architecture for handwritten text recognition systems

This paper presents an end-to-end system for reading handwritten page images. Five functional modules included in the system are introduced in this paper: (i) pre-processing, which concerns introducing an image representation for easy manipulation of large page images and image handling procedures using the image representation; (ii) line separation, concerning text line detection and extracting images of lines of text from a page image; (iii) word segmentation, which concerns locating word gaps and isolating words from a line of text image obtained efficiently and in an intelligent manner; (iv) word recognition, concerning handwritten word recognition algorithms; and (v) linguistic post-pro- cessing, which concerns the use of linguistic constraints to intelligently parse and recognize text. Key ideas employed in each functional module, which have been developed for dealing with the diversity of handwriting in its various aspects with a goal of system reliability and robustness, are described in this paper. Preliminary experiments show promising results in terms of speed and accuracy. Received October 30, 1998 / Revised January 15, 1999     

An investigation of the modified direction feature for cursive character recognition

This paper describes and analyses the performance of a novel feature extraction technique for the recognition of segmented/cursive characters that may be used in the context of a segmentation-based handwritten word recognition system. The modified direction feature (MDF) extraction technique builds upon the direction feature (DF) technique proposed previously that extracts direction information from the structure of character contours. This principal was extended so that the direction information is integrated with a technique for detecting transitions between background and foreground pixels in the character image.In order to improve on the DF extraction technique, a number of modifications were undertaken. With a view to describe the character contour more effectively, a re-design of the direction number determination technique was performed. Also, an additional global feature was introduced to improve the recognition accuracy for those characters that were most frequently confused with patterns of similar appearance. MDF was tested using a neural network-based classifier and compared to the DF and transition feature (TF) extraction techniques. MDF outperformed both DF and TF techniques using a benchmark dataset and compared favourably with the top results in the literature. A recognition accuracy of above 89% is reported on characters from the CEDAR dataset.     

Handwritten character recognition through two-stage foreground sub-sampling

In this paper, we present a methodology for off-line handwritten character recognition. The proposed methodology relies on a new feature extraction technique based on recursive subdivisions of the character image so that the resulting sub-images at each iteration have balanced (approximately equal) numbers of foreground pixels, as far as this is possible. Feature extraction is followed by a two-stage classification scheme based on the level of granularity of the feature extraction method. Classes with high values in the confusion matrix are merged at a certain level and for each group of merged classes, granularity features from the level that best distinguishes them are employed. Two handwritten character databases (CEDAR and CIL) as well as two handwritten digit databases (MNIST and CEDAR) were used in order to demonstrate the effectiveness of the proposed technique. The recognition result achieved, in comparison to the ones reported in the literature, is the highest for the well-known CEDAR Character Database (94.73%) and among the best for the MNIST Database (99.03%)     

Combining additive input noise annealing and pattern transformations for improved handwritten character recognition

Two problems that burden the learning process of Artificial Neural Networks with Back Propagation are the need of building a full and representative learning data set, and the avoidance of stalling in local minima. Both problems seem to be closely related when working with the handwritten digits contained in the MNIST dataset. Using a modest sized ANN, the proposed combination of input data transformations enables the achievement of a test error as low as 0.43%, which is up to standard compared to other more complex neural architectures like Convolutional or Deep Neural Networks.     

Recognition of handwritten numerals by Quantum Neural Network with fuzzy features

This paper describes a new kind of neural network – Quantum Neural Network (QNN) – and its application to the recognition of handwritten numerals. QNN combines the advantages of neural modelling and fuzzy theoretic principles. Novel experiments have been designed for in-depth studies of applying the QNN to both real data and confusing images synthesized by morphing. Tests on synthesized data examine QNN's fuzzy decision boundary with the intention to illustrate its mechanism and characteristics, while studies on real data prove its great potential as a handwritten numeral classifier and the special role it plays in multi-expert systems. An effective decision-fusion system is proposed and a high reliability of 99.10% has been achieved. Received October 26, 1998 / Revised January 9, 1999     

Eigen-deformations for elastic matching based handwritten character recognition

Deformations in handwritten characters have category-dependent tendencies. In this paper, the estimation and the utilization of such tendencies called eigen-deformations are investigated for the better performance of elastic matching based handwritten character recognition. The eigen-deformations are estimated by the principal component analysis of actual deformations automatically collected by the elastic matching. From experimental results it was shown that typical deformations of each category can be extracted as the eigen-deformations. It was also shown that the recognition performance can be improved significantly by using the eigen-deformations for the detection of overfitting, which is the main cause of the misrecognition in the elastic matching based recognition methods.     

Combining multiple classifiers based on third-order dependency for handwritten numeral recognition

Storing and estimating high order probability distribution of classifiers and class labels is exponentially complex and unmanageable, so we rely on an approximation scheme using the dependency. As an extension of the second-order dependency approach, the probability distribution is optimally approximated by the third-order dependency and then multiple classifiers are combined by such third-order dependency approximation.     

SVM-based hierarchical architectures for handwritten Bangla character recognition

We propose support vector machine (SVM) based hierarchical classification schemes for recognition of handwritten Bangla characters. A comparative study is made among multilayer perceptron, radial basis function network and SVM classifier for this 45 class recognition problem. SVM classifier is found to outperform the other classifiers. A fusion scheme using the three classifiers is proposed which is marginally better than SVM classifier. It is observed that there are groups of characters having similar shapes. These groups are determined in two different ways on the basis of the confusion matrix obtained from SVM classifier. In the former, the groups are disjoint while they are overlapped in the latter. Another grouping scheme is proposed based on the confusion matrix obtained from neural gas algorithm. Groups are disjoint here. Three different two-stage hierarchical learning architectures (HLAs) are proposed using the three grouping schemes. An unknown character image is classified into a group in the first stage. The second stage recognizes the class within this group. Performances of the HLA schemes are found to be better than single stage classification schemes. The HLA scheme with overlapped groups outperforms the other two HLA schemes.     

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.Received: 17 September 2002, Accepted: 22 October 2002, Published online: 4 July 2003     

A hybrid method for robust car plate character recognition

Image-based car plate recognition is an indispensable part of an intelligent traffic system. The quality of the images taken for car plates, especially for Chinese car plates, however, may sometimes be very poor, due to the operating conditions and distortion because of poor photographical environments. Furthermore, there exist some “similar” characters, such as “8” and “B”, “7” and “T” and so on. They are less distinguishable because of noises and/or distortions. To achieve robust and high recognition performance, in this paper, a two-stage hybrid recognition system combining statistical and structural recognition methods is proposed. Car plate images are skew corrected and normalized before recognition. In the first stage, four statistical sub-classifiers recognize the input character independently, and the recognition results are combined using the Bayes method. If the output of the first stage contains characters that belong to prescribed sets of similarity characters, structure recognition method is used to further classify these character images: they are preprocessed once more, structure features are obtained from them and these structure features are fed into a decision tree classifier. Finally, genetic algorithm is employed to achieve optimum system parameters. Experiments show that our recognition system is very efficient and robust. As part of an intelligent traffic system, the system has been in successful commercial use.     

Maximum mutual information training for an online neural predictive handwritten word recognition system

In this paper, we present a hybrid online handwriting recognition system based on hidden Markov models (HMMs). It is devoted to word recognition using large vocabularies. An adaptive segmentation of words into letters is integrated with recognition, and is at the heart of the training phase. A word-model is a left-right HMM in which each state is a predictive multilayer perceptron that performs local regression on the drawing (i.e., the written word) relying on a context of observations. A discriminative training paradigm related to maximum mutual information is used, and its potential is shown on a database of 9,781 words. Received June 19, 2000 / Revised October 16, 2000     

Experiments with adaptation strategies for a prototype-based recognition system for isolated handwritten characters

This paper describes an adaptive recognition system for isolated handwritten characters and the experiments carried out with it. The characters used in our experiments are alphanumeric characters, including both the upper- and lower-case versions of the Latin alphabets and three Scandinavian diacriticals. The writers are allowed to use their own natural style of writing. The recognition system is based on the k-nearest neighbor rule. The six character similarity measures applied by the system are all based on dynamic time warping. The aim of the first experiments is to choose the best combination of the simple preprocessing and normalization operations and the dissimilarity measure for a multi-writer system. However, the main focus of the work is on online adaptation. The purpose of the adaptations is to turn a writer-independent system into writer-dependent and increase recognition performance. The adaptation is carried out by modifying the prototype set of the classifier according to its recognition performance and the user's writing style. The ways of adaptation include: (1) adding new prototypes; (2) inactivating confusing prototypes; and (3) reshaping existing prototypes. The reshaping algorithm is based on the Learning Vector Quantization. Four different adaptation strategies, according to which the modifications of the prototype set are performed, have been studied both offline and online. Adaptation is carried out in a self-supervised fashion during normal use and thus remains unnoticed by the user. Received June 30, 1999 / Revised September 29, 2000     

Online handwriting recognition: the NPen++ recognizer

This paper presents the online handwriting recognition system NPen++ developed at the University of Karlsruhe and Carnegie Mellon University. The NPen++ recognition engine is based on a multi-state time delay neural network and yields recognition rates from 96% for a 5,000 word dictionary to 93.4% on a 20,000 word dictionary and 91.2% for a 50,000 word dictionary. The proposed tree search and pruning technique reduces the search space considerably without losing too much recognition performance compared to an exhaustive search. This enables the NPen++ recognizer to be run in real-time with large dictionaries. Initial recognition rates for whole sentences are promising and show that the MS-TDNN architecture is suited to recognizing handwritten data ranging from single characters to whole sentences. Received September 3, 2000 / Revised October 9, 2000     

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 digit recognition: benchmarking of state-of-the-art techniques

This paper presents the results of handwritten digit recognition on well-known image databases using state-of-the-art feature extraction and classification techniques. The tested databases are CENPARMI, CEDAR, and MNIST. On the test data set of each database, 80 recognition accuracies are given by combining eight classifiers with ten feature vectors. The features include chaincode feature, gradient feature, profile structure feature, and peripheral direction contributivity. The gradient feature is extracted from either binary image or gray-scale image. The classifiers include the k-nearest neighbor classifier, three neural classifiers, a learning vector quantization classifier, a discriminative learning quadratic discriminant function (DLQDF) classifier, and two support vector classifiers (SVCs). All the classifiers and feature vectors give high recognition accuracies. Relatively, the chaincode feature and the gradient feature show advantage over other features, and the profile structure feature shows efficiency as a complementary feature. The SVC with RBF kernel (SVC-rbf) gives the highest accuracy in most cases but is extremely expensive in storage and computation. Among the non-SV classifiers, the polynomial classifier and DLQDF give the highest accuracies. The results of non-SV classifiers are competitive to the best ones previously reported on the same databases.     

Zoning methods for handwritten character recognition: A survey

This paper presents a survey on zoning methods for handwritten character recognition. Through the analysis of the relevant literature in the field, the most valuable zoning methods are presented in terms of both topologies and membership functions. Throughout the paper, diverse zoning topologies are presented based on both static and adaptive approaches. Concerning static approaches, uniform and non-uniform zoning strategies are discussed. When adaptive zonings are considered, manual and automatic strategies for optimal zoning design are illustrated as well as the most appropriate zoning representation techniques. In addition, the role of membership functions for zoning-based classification is highlighted and the diverse approaches to membership function selection are presented. Concerning global membership functions, the paper introduces order-based approaches as well as fuzzy approaches using border-based and ranked-based fuzzy membership values. Concerning local membership functions, the recent parameter-based approaches are described, in which the optimal membership-function is selected for each zone of the zoning method. Finally, a comparative analysis on the performance of zoning methods is presented and the most interesting approaches are focused on in terms of topology design and membership function selection. A list of selected references is provided as a useful tool for interested researchers working in the field.     

Robust vision-based features and classification schemes for off-line handwritten digit recognition

We use well-established results in biological vision to construct a model for handwritten digit recognition. We show empirically that the features extracted by our model are linearly separable over a large training set (MNIST). Using only a linear discriminant system on these features, our model is relatively simple yet outperforms other models on the same data set. In particular, the best result is obtained by applying triowise linear support vector machines with soft voting on vision-based features extracted from deslanted images.     

A genetic framework using contextual knowledge for segmentation and recognition of handwritten numeral strings

For the first time, a genetic framework using contextual knowledge is proposed for segmentation and recognition of unconstrained handwritten numeral strings. New algorithms have been developed to locate feature points on the string image, and to generate possible segmentation hypotheses. A genetic representation scheme is utilized to show the space of all segmentation hypotheses (chromosomes). For the evaluation of segmentation hypotheses, a novel evaluation scheme is introduced, in order to improve the outlier resistance of the system. Our genetic algorithm tries to search and evolve the population of segmentation hypotheses, and to find the one with the highest segmentation/recognition confidence. The NIST NSTRING SD19 and CENPARMI databases were used to evaluate the performance of our proposed method. Our experiments showed that proper use of contextual knowledge in segmentation, evaluation and search greatly improves the overall performance of the system. On average, our system was able to obtain correct recognition rates of 95.28% and 96.42% on handwritten numeral strings using neural network and support vector classifiers, respectively. These results compare favorably with the ones reported in the literature.