This paper shows noise reduction on nasal skin temperature measured by a radiation thermometer with differential revision filtering. We evaluated and examined emotion stress from the nasal skin temperature fluctuation of the space-time with the radiation thermometer. The measured data with the radiation thermometer consists of three kind of factor, because of one point [24±3φ/m] measurement system. Three kinds of factors are temperature fluctuation dependent feelings stress, noise of the radiation thermometer and the other parts which it should measure as the subject moves. We make threshold processing on a change of an area in time of a measurement signal provided in a radiation thermometer. We suggested Differential Revision Filtering (DRF) that a new method use a measurement value of change of one second and a low pass filter go through. By using this method, we got the possibility that decrease movement of a measurement and a noise of measurement device.

View full abstract

The aim of this study was to develop a new evaluation method for controlling skill of a myoelectric control hand (MCH). Four amputees who daily used the MCH participated in this study. Measurement of two signals which were needed to control motion of the MCH was made. The two signals were assigned average rectified electromyogram of forearm extensors and flexors, respectively. The values of the signals increased depending on the muscles contraction level. Therapy for using the MCH was done to improve the ability of separate contractions of two muscles. This ability was defined as the skill to control motion of the MCH. The skill in each motion was evaluated by regression coefficient which was calculated by the least squares method. Regression coefficient was in inverse proportion to the skill. The quantitative index of the skill would be provide by developed method, and should be useful for efficient therapy to improve the controlling skill of myoelectric control hand.

View full abstract

In this paper, we propose an adaptive notch filter that realizes the phase shift of the all-pass filter by the expression of Fourier sine series. The method to control the weights of the all-pass filter composed by IIR filter is known well. However, these adaptive notch filters have not been put to practical use since the peculiar stability problem of IIR filter and the problem concerning the determination method of the degree of IIR filter exist. The proposed adaptive notch filter is always stable since the all-pass filter is composed by an exponential filter. Moreover, the adaptive algorithm is a simple gradient based algorithm and the convergence of weights is guaranteed. The degree of the all-pass filter only depends on the required steepness of the elimination characteristic and does not depend on the number of narrowband signals that should be eliminated. Moreover, it can be said that the proposed method is a suitable method for practical use since the elimination characteristic hardly influenced by the broadband signal added to the narrowband signal. Finally, the analytical results and the above mentioned feature are strictly verified by computer simulation.

View full abstract

In this paper, an microwave oven for disinfectant one application has been modeled by FDTD-HTE Method and reserched uniform heating of solution model simulated solid food and high viscosity food. We suggest two technique for uniform heating. First one is that water film is arranged around the food for reducing local heating of edge heating. Second one is that food is heated by using both steam and microwave. As a result, 2mm water film lower temperature of high temperature part in the food efficiently, by using uniform heating technique using water film. And we confirmed that 33% of uneven heating is improved. Moreover, we confirmed that steam heated edge of food and microwave heated center of food, and food is gone to about 60 degrees centigrade through uniform technique using both Steam and microwave.

View full abstract

In the existing methods of a moving object detection using image recognition technology, they have processed an obtained whole image, and have detected a moving object in a picture. However, when a moving portion in the scene is hidden by some obstacles, the recognition of a moving object is sometimes difficult. In this study, we propose a novel method of discriminating moving objects, such as a person or a vehicle. We use only a narrow and tall area of the video called the Strip Frame Image for detecting moving portions. By using this method, we are able to obtain the patterns of moving objects while avoiding the obstacles in a picture. Then we classify them by DP matching against previously stored reference patterns in the database for all possible classes (person, bicycle, car, and bus).
In this paper, we compare several variations of an algorithm used to detect and classify objects passing laterally in front of a security camera. We test both a moving object speed independent and a speed based method for constructing patterns for the passing objects. The results show that the relatively simple method of pattern classification by DP matching can be successfully applied for classifying graphic objects of a certain degree of complexity. Finally, non-homogeneity of people's patterns and their subsequent frequent misclassification is addressed by not producing reference patterns for people, and differentiating them from correctly classified bicycles by the DP distance to the first candidate.

View full abstract

This paper deals with stability analysis and control design for a class of nonlinear systems with input constraint when using the T-S fuzzy model. As a result, it arrives at an feedback controller that is composed of two parts: one is obtained by solving certain linear matrix inequalities (LMIs) (fixed part) to deal with the input constraint, and another one is acquired by an adaptive law (variable part) to deal with the reconstruction error between the real system and its T-S fuzzy model. Also, a discussion on how to reduce the conservatism of LMI conditions is provided. An inverted pendulum is given to demonstrate the effectiveness of the proposed feedback controller.

View full abstract

This paper evaluates a newly proposed backtrack free path planning algorithm (BFA) for manipulators. BFA is an exact algorithm, i.e. it is resolution complete. Different from existing resolution complete algorithms, its computation time and memory space are proportional to the number of arms. Therefore paths can be calculated within practical and predetermined time even for manipulators with many arms, and it becomes possible to plan complicated motions of multi-arm manipulators in fully automated environments. The performance of BFA is evaluated for 2-dimensional environments while changing the number of arms and obstacle placements. Its performance under locus and attitude constraints is also evaluated. Evaluation results show that the computation volume of the algorithm is almost the same as the theoretical one, i.e. it increases linearly with the number of arms even in complicated environments. Moreover BFA achieves the constant performance independent of environments.

View full abstract

The present paper proposes a new reinforcement learning (RL) system called a state predictor based RL system in order to solve the explosion of state space and create cooperative behaviors in multi-agent systems. The proposed system realizes a predictive function by representing both the present and the next state-action groups with ITPM which is one of incremental topology maps. The proposed system is applied to pursuit problem, and its performance is evaluated by comparing with conventional RL method through computer simulations. The experimental result shows that the proposed system can appropriately learn in a complex environment which is hardly solved by conventional RL. Furthermore, it is confirmed that the proposed system can acquire cooperative strategies.

View full abstract

Recently, Genetic Network Programming (GNP) has been proposed, which is an extension of Genetic Algorithm(GA) and Genetic Programming(GP). GNP can make compact programs and can memorize the past history in it implicitly, because it expresses the solution by directed graphs and therefore, it can reuse the nodes. In this research, intron-like nodes are introduced for improving the performance of GNP. The aim of introducing intron-like nodes is to use every node as much as possible. It is found from simulations that the intron-like nodes are useful for improving the training speed and generalization ability.

View full abstract

Recently, the rapid growth of Internet Technology and the deployment of high speed networks have produced many kinds of multimedia services. Together with it, demands for variase information and its data have been increaded and confused. The requirement for the scale of them are rapidly changing the configuration of Internet content delivery and traffic. Preexisting contents delivery systems were designed based on the Client/Server model, which brings on problems such as frequent disconnections and low speed communication due to high load on the central server. The traditional Client/Server system to delivere contents, therefore, is in denger of attainment of limit, because Internet users have been increased explosively and the quantity and quality of the need for the contents have been enhanced. In order to solve these problems, Peer-to-Peer (P2P) networks are introduced and expanded through Internet environments. In this paper, an education contents delivery system for cyber education system using idle computing powers in the hybrid P2P computing to share computing resources is designed and implement.

View full abstract

In the ubiquitous computing environment, it is expected to provide context-aware services enabled by using physical environmental information acquired from massive embedded sensors. One of issues to provide these services is interoperability of different middleware. Among these middleware a service roaming is occurred during executing the service. This paper provides our approaches of technologies for interoperating different middleware, composed of sharing information of resources by normalization of profiles, normalization of service description and adapting original specifications which cannot be normalized. These techniques are based on the model of Super Distributed Objects (SDO) that is standardized in OMG^TM. In order to verify these techniques, we have experimented interoperation of different middleware.

View full abstract

We propose an apparel online shopping site that the fashion adviser exists on the internet. The fashion adviser, who has detailed knowledge about the fashion in real shop, selects and coordinates the clothes of the customer's preference. However, the customer, who didn't have detailed knowledge about the fashion, was not able to choose the clothes suitable for the customer's preference from among the candidate of a large amount of clothes on a conventional apparel shopping site. Then, we compose the system that analyzes the customer's preference by the AHP technique, makes to the cluster by the correlation of clothes, and analyzes the market basket. As a result, this system can coordinate the clothes appropriate for the favor of an individual customer. Moreover, this system can propose the recommendation of other clothes based on past sales data.

View full abstract

Super-Function Based Machine Translation(SFBMT) which is a type of Example-Based Machine Translation has a feature which makes it possible to expand the coverage of examples by changing nouns into variables, however, there were problems extracting entire date/time expressions containing parts-of-speech other than nouns, because only nouns/numbers were changed into variables. We describe a method for extracting date/time expressions for SFBMT. SFBMT uses noun determination rules to extract nouns and a bilingual dictionary to obtain correspondence of the extracted nouns between the source and the target languages. In this method, we add a rule to extract date/time expressions and then extract date/time expressions from a Japanese-English bilingual corpus. The evaluation results shows that the precision of this method for Japanese sentences is 96.7%, with a recall of 98.2% and the precision for English sentences is 94.7%, with a recall of 92.7%.

View full abstract

In this paper, an improved guided genetic algorithm is proposed forJob-shop schueduling problem. The proposed method is improved by genetic algorithm using multipliers which can be adjusted during the search process. The simulation results based on some benchmark problems that proves the proposed method can find better solutions than genetic algorithm and original guided genetic algorithm.

View full abstract

A paper filtering system that supports the effective collection of related technical papers is becoming important as the technological progress has been rapid. Two requirements for the paper filtering system are [A] the reduction of user's workload in specifying filtering conditions and [B] sufficient filtering accuracy. In this paper, we propose a paper filtering method that meets both [A] and [B] simultaneously by focusing on feartures of co-author research group, subject category, and terminology. The result of evaluation using 3,600 domestic society papers shows that the proposed methods improved mean average precision from 0.39 to 0.50 by 0.11 comparing with conventional pseudo relevance feedback method, which becomes closer to practical use.

View full abstract

The applications of embedded system technologies have spread widely in various products, such as home appliances, cellular phones, automobiles, industrial machines and so on. Due to intensified competition, embedded software has expanded its role in realizing sophisticated functions, and new development methods like a hardware/software (HW/SW) co-design for uniting HW and SW development have been researched. The shortfall of embedded SW engineers was estimated to be approximately 99,000 in the year 2006, in Japan. Embedded SW engineers should understand HW technologies and system architecture design as well as SW technologies. However, a few universities offer this kind of education systematically. We propose a student experiment method for learning the basics of embedded system development, which includes a set of experiments for developing embedded SW, developing embedded HW and experiencing HW/SW co-design. The co-design experiment helps students learn about the basics of embedded system architecture design and the flow of designing actual HW and SW modules. We developed these experiments and evaluated them.

View full abstract

This paper discusses phase noise characteristics of an Fr oscillator focusing on a signal output position. Oscillation circuits can be divided into an amplifier and a feedback circuit. We compared that phase noise characteristics before passing through a feedback circuit or a after. As a result, it turned out that the slope of the phase noise characteristic does not relate to the signal output position.

View full abstract

This study proposes a copy-learning method which can recognize hollow characters by only learning the standard characters. The method learns standard characters by a part of the 3-layer net, copies obtained weights and biases on an unused part of the net by a copy-learning rule, and recognizes hollow characters. We had confirmed the effectiveness of this method from several experiments.

View full abstract