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

A fast support vector machine (SVM) training algorithm is proposed under the decomposition framework of SVM's algorithm by effectively integrating kernel caching, digest and shrinking policies and stopping conditions. Extensive experiments on MNIST handwritten digit database have been conducted to show that the proposed algorithm is much faster than Keerthi et al.'s improved SMO, about 9 times. Combined with principal component analysis, the total training for ten one against the-rest classifiers on MNIST took just 0.77 hours. The promising scalability of the proposed scheme can make it possible to apply SVM to a wide variety of problems in engineering.

A Fast SVM Training Algorithm

This paper presents a new approach to document analysis. The proposed approach is based on modified fractal signature. Instead of the time-consuming traditional approaches (top-down and bottom-up approaches) where iterative operations are necessary to break a document into blocks to extract its geometric (layout) structure, this new approach can divide a document into blocks in only one step. This approach can be used to process documents with high geometrical complexity. Experiments have been conducted to prove the proposed new approach for document processing.

A new approach to document analysis based on modified fractal signature

This paper introduces the Center for Pattern Recognition and Machine Intelligence (CENPARMI) Farsi dataset which can be used to measure the performance of handwritten recognition and word spotting systems. This dataset is unique in terms of its large number of gray and binary images (432,357 each) consisting of dates, words, isolated letters, isolated digits, numeral strings, special symbols, and documents. The data was collected from 400 native Farsi writers. The selection of Farsi words has been based on their high frequency in financial documents. The dataset is divided into grouped and ungrouped subsets which will give the user the flexibility of whether or not to use CENPARMI's pre-divided dataset (60% of the images are used as the Training set, 20% of the images as the Validation set, and the rest as the Testing set). Finally, experiments have been conducted on the Farsi isolated digits with a recognition rate of 96.85%.

A New Large-Scale Multi-purpose Handwritten Farsi Database

Automatic recognition of serial numbers in bank notes

A new isolated handwritten Farsi numeral recognition algorithm is proposed in this paper, which exploits the sparse and over-complete structure from the handwritten Farsi numeral data. In this research, the sparse structure is represented as an over-complete dictionary, which is learned by the K-SVD algorithm. These atoms in this dictionary are adopted to initialize the first layer of the Convolutional Neural Network (CNN), the latter is then trained to do the classification task. Data distortion techniques are also applied to promote the generalization capability of the trained classifier. Experiments have shown that good results have been achieved in CENPARMI handwritten Farsi numeral database.

http://www.cvc.uab.es/icdar2009/papers/3725a586.pdf

Ching Y. Suen

Papers

A Fast SVM Training Algorithm

A new approach to document analysis based on modified fractal signature

A New Large-Scale Multi-purpose Handwritten Farsi Database

Automatic recognition of serial numbers in bank notes

Isolated Handwritten Farsi Numerals Recognition Using Sparse and Over-Complete Representations