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 near effective and efficient model in recognition

A new method for segmentation of unconstrained handwritten numeral strings is proposed, and its performance in segmentation of handwritten Farsi and English numeral strings is compared. Our segmentation method uses a combination of features from the foreground and background of the image. In order to find foreground features, an algorithm based on skeleton tracing is introduced for the first time. The skeletons of the connected digits are traversed in clockwise and anti- clockwise directions, and intersection points are mapped onto the outer contour to form our foreground features. In order to find background features, another new algorithm is proposed. Considering vertical projections of top and bottom profiles, two background skeletons are found, and feature points are extracted from them. Background and foreground feature points are connected together to construct segmentation paths. Finally each segmentation path is evaluated based on the properties of its left and right connected components. Experiments show our method can provide a list of good segmentation hypotheses for segmentation- based recognition systems. The results of our system are shown on both Farsi and English numeral strings.

Segmentation of Handwritten Numeral Strings in Farsi and English Languages

This paper describes a post-processing scheme to improve recognition performance of touching handwritten numeral strings. Verification factors are defined to rectify the recognition results of digits that are segmented by the highest reliability values. Three kinds of verification factors from structural features and recognition probability are used to determine mis-segmented digits. Additional segment digit set is included in valid digit sets when it satisfies the reconsideration condition. A final optimal digit set is selected as the highest ranked segment digit set among all candidate segmented digit sets. Introduction of reconsideration condition and additional segment digit set improves the reliability of the recognition result as well as recognition performance. Experiments were carried out with touching handwritten numeral strings of NIST SD19 database and an encouraging recognition result was obtained.

Post-processing scheme for improving recognition performance of touching handwritten numeral strings

Evaluation of expert systems

The current advancements in machine intelligence have expedited the process of recognizing vehicles and other objects on the roads. Several methods including Deep Learning techniques have been proposed recently for LPR, yet those methods are limited to specific regions or privately collected datasets. In this paper, we propose an end-to-end Deep Convolutional Neural Network system for license plate recognition that is not limited to a specific region or country. We apply a modified version of YOLO v2 to first recognize the vehicle and then locate the license plate. Moreover, through the convolutional procedures, we improve an Optical Character Recognition network (OCR-Net) to recognize the license plate numbers and letters. Our method performs well for different vehicle types. Our system overcomes tilted and distorted license plate images and performs adequately under various illumination conditions, and noisy backgrounds. Our experimental results on 4,837 images of stationary and moving vehicles (cars, buses, motorbikes, and trucks) from different countries show that our proposed system achieved recognition rates between 88.5% and 98.04%, outperforming the state-of-the-art commercial and academic methods for challenging images.

Ching Y. Suen

Papers

A near effective and efficient model in recognition

Segmentation of Handwritten Numeral Strings in Farsi and English Languages

Post-processing scheme for improving recognition performance of touching handwritten numeral strings

Evaluation of expert systems

License Plate Detection and Recognition by Convolutional Neural Networks