Ching Y. Suen

Methods of combining multiple classifiers and their applications to handwriting recognition

Possible solutions to the problem of combining classifiers can be divided into three categories according to the levels of information available from the various classifiers. Four approaches based on different methodologies are proposed for solving this problem. One is suitable for combining individual classifiers such as Bayesian, k-nearest-neighbor, and various distance classifiers. The other three could be used for combining any kind of individual classifiers. On applying these methods to combine several classifiers for recognizing totally unconstrained handwritten numerals, the experimental results show that the performance of individual classifiers can be improved significantly. For example, on the US zipcode database, 98.9% recognition with 0.90% substitution and 0.2% rejection can be obtained, as well as high reliability with 95% recognition, 0% substitution, and 5% rejection. >

Computer recognition of unconstrained handwritten numerals

Four independently, developed expert algorithms for recognizing unconstrained handwritten numerals are presented. All have high recognition rates. Different experimental approaches for incorporating these recognition methods into a more powerful system are also presented. The resulting multiple-expert system proves that the consensus of these methods tends to compensate for individual weaknesses, while preserving individual strengths. It is shown that it is possible to reduce the substitution rate to a desired level while maintaining a fairly high recognition rate in the classification of totally unconstrained handwritten ZIP code numerals. If reliability is of the utmost importance, substitutions can be avoided completely (reliability=100%) while retaining a recognition rate above 90%. Results are compared with those for some of the most effective numeral recognition systems found in the literature. >

Automatic recognition of handprinted characters—The state of the art

Based on a study of the extensive literature in handprint recognition, this paper presents a survey in this challenging field. Recognition algorithms, data bases, character models, and handprint standards are examined. Achievements in the recognition of handprinted numerals, alphanumerics, Fortran, and Katakana characters are analyzed and compared. Data quality and constraints, as well as human and machine factors are also described. Characteristics, problems, and actual results on on-line recognition of handprinted characters for different applications are discussed. New emphases and directions are suggested.

Optimal combinations of pattern classifiers

To improve recognition results, decisions of multiple classifiers can be combined. We study the performance of combination methods that are variations of the majority vote. A Bayesian formulation and a weighted majority vote (with weights obtained through a genetic algorithm) are implemented, and the combined performances of 7 classifiers on a large set of handwritten numerals are analyzed.

A recursive thresholding technique for image segmentation

In this correspondence, we present a general recursive approach for image segmentation by extending Otsus (1978) method. The new approach has been implemented in the scope of document images, specifically real-life bank checks. This approach segments the brightest homogeneous object from a given image at each recursion, leaving only the darkest homogeneous object after the last recursion. The major steps of the new technique and the experimental results that illustrate the importance and the usefulness of the new approach for the specified class of document images of bank checks will be presented.

Fast SVM training algorithm with decomposition on very large data sets

Training a support vector machine on a data set of huge size with thousands of classes is a challenging problem. This paper proposes an efficient algorithm to solve this problem. The key idea is to introduce a parallel optimization step to quickly remove most of the nonsupport vectors, where block diagonal matrices are used to approximate the original kernel matrix so that the original problem can be split into hundreds of subproblems which can be solved more efficiently. In addition, some effective strategies such as kernel caching and efficient computation of kernel matrix are integrated to speed up the training process. Our analysis of the proposed algorithm shows that its time complexity grows linearly with the number of classes and size of the data set. In the experiments, many appealing properties of the proposed algorithm have been investigated and the results show that the proposed algorithm has a much better scaling capability than Libsvm, SVM/sup light/, and SVMTorch. Moreover, the good generalization performances on several large databases have also been achieved.

An HMM-based approach for off-line unconstrained handwritten word modeling and recognition

Describes a hidden Markov model-based approach designed to recognize off-line unconstrained handwritten words for large vocabularies. After preprocessing, a word image is segmented into letters or pseudoletters and represented by two feature sequences of equal length, each consisting of an alternating sequence of shape-symbols and segmentation-symbols, which are both explicitly modeled. The word model is made up of the concatenation of appropriate letter models consisting of elementary HMMs and an HMM-based interpolation technique is used to optimally combine the two feature sets. Two rejection mechanisms are considered depending on whether or not the word image is guaranteed to belong to the lexicon. Experiments carried out on real-life data show that the proposed approach can be successfully used for handwritten word recognition.

Developing knowledge management metrics for measuring intellectual capital

Measuring intellectual capital is a growing area of interest in the knowledge management field. Metrics are being developed and applied by some organizations, but there needs to be more research throughout the international community to better define these measures. One limitation of the current measures is that they do not necessarily address the “knowledge level” and the types of value‐added knowledge that individuals obtain. This paper takes a look at the current measures, discusses some possible limitations, and suggests some additional measures that could be used in the intellectual capital area to complement existing measures.

Automatic recognition of handwritten numerical strings: a recognition and verification strategy

A modular system to recognize handwritten numerical strings is proposed. It uses a segmentation-based recognition approach and a recognition and verification strategy. The approach combines the outputs from different levels such as segmentation, recognition, and postprocessing in a probabilistic model. A new verification scheme which contains two verifiers to deal with the problems of oversegmentation and undersegmentation is presented. A new feature set is also introduced to feed the oversegmentation verifier. A postprocessor based on a deterministic automaton is used and the global decision module makes an accept/reject decision. Finally, experimental results on two databases are presented: numerical amounts on Brazilian bank checks and NIST SD19. The latter aims at validating the concept of modular system and showing the robustness of the system using a well-known database.

A trainable feature extractor for handwritten digit recognition

This article focuses on the problems of feature extraction and the recognition of handwritten digits. A trainable feature extractor based on the LeNet5 convolutional neural network architecture is introduced to solve the first problem in a black box scheme without prior knowledge on the data. The classification task is performed by support vector machines to enhance the generalization ability of LeNet5. In order to increase the recognition rate, new training samples are generated by affine transformations and elastic distortions. Experiments are performed on the well-known MNIST database to validate the method and the results show that the system can outperform both SVMs and LeNet5 while providing performances comparable to the best performance on this database. Moreover, an analysis of the errors is conducted to discuss possible means of enhancement and their limitations.