Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.

Gradient-based learning applied to document recognition

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques and methods imported from statistical learning theory have been receiving increasing attention. The design of a recognition system requires careful attention to the following issues: definition of pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design and learning, selection of training and test samples, and performance evaluation. In spite of almost 50 years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition, and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.

/pdf/statistical-pattern-recognition-a-review-1kz1b0tte4.pdf

Statistical pattern recognition: a review

We develop a common theoretical framework for combining classifiers which use distinct pattern representations and show that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision. An experimental comparison of various classifier combination schemes demonstrates that the combination rule developed under the most restrictive assumptions-the sum rule-outperforms other classifier combinations schemes. A sensitivity analysis of the various schemes to estimation errors is carried out to show that this finding can be justified theoretically.

On combining classifiers

From the Publisher:
Pattern recognition has long been studied in relation to many different (and mainly unrelated) applications, such as remote sensing, computer vision, space research, and medical imaging. In this book Professor Ripley brings together two crucial ideas in pattern recognition; statistical methods and machine learning via neural networks. Unifying principles are brought to the fore, and the author gives an overview of the state of the subject. Many examples are included to illustrate real problems in pattern recognition and how to overcome them.This is a self-contained account, ideal both as an introduction for non-specialists readers, and also as a handbook for the more expert reader.

Pattern recognition and neural networks

Removal of noise patterns in handwritten images using expectation maximization and fuzzy inference systems

Describes the latest developments to enhance the performance of our HMM-based handwritten word recognition system. These methods only deal with the recognition phase and involve the improvement of the HMM architecture as well as the optimization of the training phase. Experiments carried out on real data show that the proposed approaches lead to significant improvements in the accuracy of the system.

Improved model architecture and training phase in an off-line HMM-based word recognition system

Classification finds wide applications in artificial intelligence and expert systems. Feature extraction is a key step for classifier learning. However, the relation among samples is usually ignored in classical feature extraction models. Recently, feature extraction based on graph signal processing that makes use of the relation among samples has attracted great attention. It is a common assumption that the classification information is smooth and of low frequency in these studies. We point out that it is the discrimination ability that essentially makes a good classification feature instead of smoothness. This new perspective prompts us to introduce the concept of discriminative graph signal, and then, based on this concept, we propose a novel feature extraction model for supervised classification. To improve the classification ability for multi-class problems, a generalized model is proposed to extract multiple discriminative signals and an algorithm is also presented to compute the multiple discriminative signals simultaneously. On five publicly available UCI datasets, our proposed method outperforms the existing methods in terms of performance. Finally some drawbacks are discussed and future research directions are also provided.

A feature extraction model based on discriminative graph signals

The authors have extended existing methods to identify the language of an on-line document after the characters have been coded using 10 character classes based on visual characteristics. In particular, they exploit word bigrams and trigrams in both a linear combination of score values and an expert systems approach. Knowledge about each language as acquired from a large number of on-line texts. Using a small set of rules, the expert system outperforms the linear combination in accuracy and shows more stability when parameter settings are varied.

Language identification of on-line documents using word shapes

In this study we evaluate different HMM topologies in terms of recognition of handwritten numeral strings by considering the framework of the Level Building Algorithm (LBA). By including an end-state in a left-to-right HMM structure we observe a significant improvement in the string recognition performance since it provides a better definition of the segmentation cuts by the LBA. In addition, this end-state allows us the use of a two-step training mechanism with the objective of integrating handwriting-specific knowledge into the numeral models to obtain a more accurate representation of numeral strings. The contextual information regarding the interaction between adjacent numerals in strings (spaces, overlapping and touching) is modeled in a pause model built into the numeral HMMs. This has shown to be a promising approach even though it is really dependent on the training database.

/pdf/an-enhanced-hmm-topology-in-an-lba-framework-for-the-3an1fj2h74.pdf

Ching Y. Suen

Papers

Removal of noise patterns in handwritten images using expectation maximization and fuzzy inference systems

Improved model architecture and training phase in an off-line HMM-based word recognition system

A feature extraction model based on discriminative graph signals

Language identification of on-line documents using word shapes

An Enhanced HMM Topology in an LBA Framework for the Recognition of Handwritten Numeral Strings