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

When linear support vector machines (SVMs) are applied to multi-class text categorization in industry, the size of the linear SVM model is very large, usually greater than several gigabytes. As a result, the model cannot directly fit into the computer memory and the classification process is slow. In this paper, a novel method based on vector norm is proposed to shrink the model size significantly without sacrificing the classification accuracy. Also, we propose a cache-efficient implementation of multi-class linear SVMs in the classification phase. Our experimental results have shown that on Yahoo-Korea dataset the proposed method can shrink the model size from 5.2 gigabytes to 260 megabytes and the efficient implementation of linear SVM has obtained a speedup factor of 44.

Effective shrinkage of large multi-class linear svm models for text categorization

An Ensemble of Character Features and Fine-Tuned Convolutional Neural Network for Spurious Coin Detection

This paper presents a new perception-based model for reading cursive script. We limit the scope of our study to the oo-line recognition of isolated cursive words. Starting from observations and assumptions used in the elaboration of reading models, we describe the organization of our pseudo-neuronal system and show the role of the activation mechanism in perceiving and reading cursive script. We have introduced into our model some characteristics speciic to cursive script. First, we use appropriate features such as ascenders, descenders and loops. Second, we deal with the ambiguity of letter location by introducing the concept of fuzzy position. The location as well as the missing letters are deduced from the context (i.e. the word-letter lexicon). After implementation of our method, preliminary qualitative results have been obtained and are discussed. We are concentrating now on further formalizing and generalizing the proposed model on a larger data base.

Using reading models for cursive script recognition

Document Script Identification (DSI) is a very useful application in document processing. This paper presents a method for this application that uses a new noise tolerant feature, the Downgraded Pixel Density feature. Compared to other features widely used in existing DSI solutions, this new feature is much more robust to variations in slant, font and style of printed documents. Experimental results show that the method achieves promising identification performances.

Noise Tolerant Script Identification of Printed Oriental and English Documents Using a Downgraded Pixel Density Feature

This paper describes a Chinese cheque processing system currently under development at the Centre for Pattern Recognition and Machine Intelligence (CENPARMI). The information on Chinese bank cheques is not the same as that on alphanumeric bank cheques. The legal amount in a Chinese bank cheque is the Chinese character text associated with each currency unit. This paper discusses a technique using each currency unit as a key word to locate/extract the legal amount in bank cheques. In the analysis and recognition process, the system tries to locate the smallest currency units in the image and identifies it first. Then, the system tries to locate the image strings associated with each currency unit. Each image string is separated and recognized. Next, a set of rules and context are applied to recognize the characters. In order to choose the correct one, the recognized character string is accepted only if it satisfies all the conditions governed by rules.

Ching Y. Suen

Papers

Effective shrinkage of large multi-class linear svm models for text categorization

An Ensemble of Character Features and Fine-Tuned Convolutional Neural Network for Spurious Coin Detection

Using reading models for cursive script recognition

Noise Tolerant Script Identification of Printed Oriental and English Documents Using a Downgraded Pixel Density Feature

Location and recognition of legal amounts on Chinese bank cheques