Toshihisa Kosaka

Rotation-invariant neural pattern recognition system with application to coin recognition

In pattern recognition, it is often necessary to deal with problems to classify a transformed pattern. A neural pattern recognition system which is insensitive to rotation of input pattern by various degrees is proposed. The system consists of a fixed invariance network with many slabs and a trainable multilayered network. The system was used in a rotation-invariant coin recognition problem to distinguish between a 500 yen coin and a 500 won coin. The results show that the approach works well for variable rotation pattern recognition.

Bill classification by using the LVQ method

For pattern classification problems the neuro-pattern recognition, which is the pattern recognition based on the neural network approach, has been increasingly popular since it can classify various patterns similar to human beings. In this paper we adopt the learning vector quantization (LVQ) method to classify the various bank notes. The reasons to use LVQ are that it can process the unsupervised classification and treat many input data with small computational burdens. We construct the LVQ network to classify the Italian Liras. Compared with a conventional pattern matching technique, which has been adopted as a classification method, the proposed method has shown excellent classification results.

Intelligent Electronic Nose Systems for Fire Detection Systems Based on Neural Networks

In this paper, an intelligent electronic nose (EN)system designed using cheap metal oxide gas sensors (MOGS) is designed to detect fires at an early stage. The time series signals obtained from the same source of fire are highly correlated, and different sources of fire exhibit unique patterns in the time series data. Therefore, the error back propagation (BP) method can be effectively used for the classification of the tested smell. The accuracy of 99.6% is achieved by using only a single training dataset from each source of fire. The accuracy achieved with the k-means algorithm is 98.3%, which also shows the high ability of the EN in detecting the early stage of fire from various sources.

Improvement of reliability in banknote classification using reject option and local PCA

The improvement of reliability in banknote neuro-classifier is investigated and a reject option is proposed based on the probability density function of the input data. The classification reliability is evaluated through two reliability parameters, which are associated with the winning class probability and the second maximal probability. Then a threshold value is set up to reject the unreliable classifications. As for modeling the non-linear correlation among the data variables and extracting the features, a local principal components analysis (PCA) is applied. The method is tested with a learning vector quantization (LVQ) classifier using 1440 data samples of various US dollar bills. The results show that by taking a suitable reject threshold value and also a proper number of regions for the local PCA, the reliability of system can be improved significantly.

A reliable method for classification of bank notes using artificial neural networks

We present a method based on principal component analysis (PCA) for increasing the reliability of bank note recognition machines. The system is intended for classifying any kind of currency, but in this paper we examine only US dollars (six different bill types). The data was acquired through an advanced line sensor, and after preprocessing, the PCA algorithm was used to extract the main features of data and to reduce the data size. A linear vector quantization (LVQ) network was applied as the main classifier of the system. By defining a new method for validating the reliability, we evaluated the reliability of the system for 1200 test samples. The results show that the reliability is increased up to 95% when the number of PCA components as well as the number of LVQ codebook vectors are taken properly. In order to compare the results of classification, we also applied hidden Markov models (HMMs) as an alternative classifier.

PID Control of Speed and Torque of Electric Vehicle

Abstract—In this paper we consider the neuro-control method and its application to control problems of an electric vehicle. The neuro-control methods adopted here is based on proportional-plus-integral-plus-derivative (PID) control, which has been adopted to solve process control or intelligent control problems. In Japan about eighty four per cent of the process industries have used the PID control. After deriving the selftuning PID control scheme (neuro-PID) using the learning ability of the neural network, we will show the control results by using the speed and torque control of an electric vehicle.

Bill money classification by competitive learning

The progress of computer science enables us to process complex and large scale computations and advanced pattern recognition methods can be adopted for pattern classification problems. Among them neuro-pattern recognition, which means pattern recognition based on neural networks, has been given attention since it has classified various patterns like human beings. We adopt the learning vector quantization (LVQ) method to classify money. The reasons for using the LVQ are that it can process unsupervised classification data and treat a large amount of input data with a small computational burden. We construct the LVQ network to classify Italian Lira. Compared with a conventional pattern matching technique, which has been adopted as a classification method, the proposed method has shown excellent classification results.

Neuro-classification of fatigued bill based on tensional acoustic signal

In the practical use of automated teller machines (ATMs), dealing with much fatigued bills causes serious trouble. To avoid this problem, rapid development of automatic classification methods that can be implemented on banking machines is desired. We propose a new automatic classification method of fatigued bill based on acoustic signal feature of a banking machine. Feeding a bill to a banking machine, a typical acoustic signal is emitted in the transportation part of the machine by tensioning the slackness of the bill transportation. The proposed method focuses on the fact that the tensional acoustic signal features differ in fatigue level of the bill, and uses spectral information of the tensional acoustic signal as the feature for classification of fatigued bill. The proposed method also uses the self organizing map (SOM) type neural network as the classifier to get high classification performance. Simulation results by using real tensional acoustic signal show the effectiveness of the proposed method.

Classification of bill fatigue levels by feature-selected acoustic energy pattern using competitive neural network

This paper proposes a new method to classify bills into different fatigue levels. Feature-selected acoustic energy patterns obtained from an acoustic signal generated by a bill passing through a banking machine are used for classification. The feature-selected acoustic energy patterns are fed to a competitive neural network with the learning vector quantization algorithm, and classified the bill into three fatigue levels. Furthermore, the selection of features in an acoustic energy pattern is performed to improve classification performance. We introduce a genetic algorithm to obtain the optimal feature selection. The experimental results show that the proposed method is useful for classification of fatigue levels of bills, and the classification performances are improved by selecting feature with genetic algorithm.

Continuous fatigue level estimation for the classification of fatigued bills based on an acoustic signal feature by a supervised SOM

Fatigued bills have a harmful influence on the daily operation of automated teller machines (ATMs). To make the classification of fatigued bills more efficient, the development of an automatic fatigued bill classification method with a continu ous fatigue level is desirable. We propose a new method to estimate the bending rigidity of bills using the acoustic signal feature of banking machines. The estimated bending rigidities are used as the continuous fatigue level for the classification of fatigued bills. By using a supervised self-organizing map (SOM), we effectively estimate the bending rigidity using only the acoustic energy pattern. The experimental results with real bill samples show the effectiveness of the proposed method.