本記事に「抄録」はありません。

抄録全体を表示

In this paper, the authors propose a method for thresholding a gray-scaled image. In the method, a threshold for a coordinate is set by an iteration that is composed of the pixels detection from neighboring area and a calculation of a candidate threshold using the detected pixels. The candidate threshold settles down to a value, because the detected pixels finally become invariable. Therefore, the method can set the threshold which follows up a light changing and a contrast changing. The proposed method and the conventional methods were tested, using a preceding car images of the inside and outside of a tunnel. We compared the method with other methods by an estimation based on a correlation on the thresholding. The result showed that the proposed method was stable relatively and set the good threshold continuously.

抄録全体を表示

Recently, the importance of understanding the number of people and the flow of the persons at public accommodation or department stores have increased more and more. This information is useful for congestion reducing, efficient promotion of the institution management and sales improvement, etc. The conventional methods of counting number of people are carried out by human viewing and by a machine of rotary stick-type counter. Therefore, we have already proposed an automatic system for counting number of people by the image processing to use a straight measurement line and a space-time image. However, these methods are not suitable for the counting at a wide place. In this paper, we propose a method of counting multidirection passer-by by using circular space-time image. In this method, a circular measurment line is set on a sequence of the background subtraction images. All pixels on this line is transformed to the space-time image. The number of passer-by can be counted by using this space-time image. But the direction information of passer-by cannot be obtained from this space-time image. Therefore, two circular measurment lines are set on a sequence of the background subtraction images. Two space-time images are generated from the outside line and the inside line. The directions of passer-by can be obtained by detecting which line passer-by passed previously.

抄録全体を表示

The La_0.5Sr_0.5CoO₃ was used as a cathode for the first time when the plasma display panel was used on the handheld personal computer. It has favorable sputtering resistance characteristic and stable discharge characteristic in gases. The La_0.5Sr_0.5CoO₃ cathode is used as a powder, and a mixture powder of La_0.5Sr_0.5CoO₃ and insulating material can be used as a cathode. When a mixture powder of La_0.5Sr_0.5CoO₃ and aluminum oxide is used as a cathode, the luminescence and the discharge voltage of the unit cell increase depending on the weight of aluminum oxide. In this report, the luminescence characteristics of the La_0.5Sr_0.5CoO₃ cathode with the addition of aluminum oxide was investigated through an analysis of the DC plasma panel which was made as prototype.
In order to find out this mechanism, the electric fields on the cathode surfaces were estimated on the mixture powder cathodes of individual weight of aluminum oxide. The electric fields on the cathode surfaces increase depending on the weight of aluminum oxide, and the luminescence increase depending on the electric fields on the cathode surfaces.

抄録全体を表示

The La_0.5Sr_0.5CoO₃ was used as a cathode for the first time when the plasma display panel was used on the handheld personal computer. It has favorable sputtering resistance characteristic and stable discharge characteristic in gases. A cathode made from the La_0.5Sr_0.5CoO₃ has unique electrical characteristic in period of continual discharge. The discharge voltage and luminance of the La_0.5Sr_0.5CoO₃ shift during an extended period of discharge, and the discharge voltage continue to increase in later period of continual discharge. It is presumed that the discharge voltage and luminance shift is caused by the discharge and luminescence area shift of the cathode surface, and that the current density shift is caused by the discharge area shrinkage phenomenon.
This discharge area shrinkage phenomenon is reduced by use of a mixture of La_0.5Sr_0.5CoO₃ and insulator, and the sputter rate is reduced, too. In this report, the discharge and luminescence area shift of the La_0.5Sr_0.5CoO₃ are investigated through an analysis of the DC plasma panel that is made as prototype, and the discharge characteristic of the La_0.5Sr_0.5CoO₃ cathode is compared with the cathode that is made from a mixture of La_0.5Sr_0.5CoO₃ and insulator.

抄録全体を表示

This paper describes ultra-flattened chromatic dispersion and low confinement loss that can be achieved from a two defected rings square photonic crystal fiber (SPCF). An efficient full vector modal solver based on the finite difference method with anisotropic perfectly matched layers absorbing boundary condition is used for numerical analysis. The designed two defected rings index-guiding SPCF has a ultra-flattened dispersion of 0 ± 0.5 ps/(nm.km) in a wavelength range of 1.42 to 1.66 μm (240 nm band) which covers S+C+L telecommunication bands and extremely low confinement loss of less than 10^-13 dB/m in a wavelength range of 1.3 to 1.7 μm. The proposed two defected rings index-guiding SPCF also shows promising dispersion tolerance and extremely low bending loss.

抄録全体を表示

This paper describes amplification characteristics of the Er-doped Ti-diffused LiNbO₃ waveguide light amplifiers made by Yb co-dope technique. For the purpose of easily analyzing, approximated solutions of the rate equations describing the transitions between the dopant populations in the energy levels in the coupled Er-Yb system are presented, which are obtained by using a perturbation method. Advantages of the amplifiers made by co-dope of Yb are demonstrated by numerical simulations. Simulation results in the high-index clad loaded configuration, in which high gain is achieved, are also demonstrated in comparisons with the conventional non-clad configuration.

抄録全体を表示

Impulse based ultra-wideband radio has lately attracted considerable attention as medical monitoring sensor since it is expected to measure bio-signals of a patient on a bed such as respiration rate and heartbeat with a remote non-contact approach. It is also friendly to the environment including the human body due to the very low electromagnetic energy emission. Using conventional ranging scheme, however, high speed A/D device should be required in order to detect the small respiratory displacement. This paper suggests a respiratory monitoring scheme where the respiration rate is measured by observing the variation of the path strength from the patient. Therefore, it does not require high speed A/D. It also makes possible to design the simultaneous monitoring of multiple patients in hospital beds, for example. In this paper the measurements were conducted for various scenarios and the feasibility is discussed.

抄録全体を表示

This paper describes a 3-D motion capture system for the quantitative evaluation of a finger-nose test using image processing. In the field of clinical medicine, qualitative and quantitative evaluation of voluntary movements is necessary for correct diagnosis of disorders. For this purpose, we have developed a 3-D measuring system with a multi-camera system. The configuration of the system is described and examples of movement data are shown for normal subjects and patients. In the finger-nose test at a fast trial speed, a discriminant analysis using Maharanobis generalized distances shows a discriminant rate of 93% between normal subjects and spinocerebellar degeneration(SCD) patients.

抄録全体を表示

A goal of this paper is to construct the adaptive output tracking controller for nonlinear multivariable systems with unknown parameters. For this purpose, an interactor for nonlinear systems is used as a prior information of the unknown plant. If the interactor is diagonal, the problem becomes straightforward extension of the adaptive control problem for scalar systems. However, as in the linear case, the interactor is a lower left triangular polynomial matrix whose off-diagonal elements depend on the unknown parameters of the plant. In this paper, first, using the interactor algorithm, the structure of the interactor of the unknown plant is estimated, and then, the well developed MRACS technique for linear systems will be applied. Some simulation results will be presented to show the validity of this designe concept. This shows that the framework for designing the MRACS for linear multivariable systems can be also used for the nonlinear multivariable system.

抄録全体を表示

This paper describes a novel operationality improvement control for electric power assisted wheelchairs. “Electric power assisted wheelchair” which assists the driving force by electric motors is expected to be widely used as a mobility support system for elderly people and disabled people, however, the performance of the straight and circular road driving must be further improved because the two wheels drive independently. This paper proposes a novel operationality improvement control by fuzzy algorithm to realize the stable driving on straight and circular roads. The suitable assisted torque of the right and left wheels is determined by fuzzy algorithm based on the posture angular velocity, the posture angle of the wheelchair, the human input torque proportion and the total human torque of the right and left wheels. Some experiments on the practical roads show the effectiveness of the proposed control system.

抄録全体を表示

The present paper proposes a detection method of inclined cracks in concrete structures using an acceleration pickup. The sensor on the concrete surface observes the stationary waves inside the structure whose frequencies depend on the distance to the crack (or to the bottom of the structure). The method can recognize the profile of the inclined crack with a high precision by the following two procedures. The first is to model the sensor output as an output of a linear dynamic system with an unknown parameter and measures the accurate propagation distance from the sensor to the crack by applying a maximum likelihood method at each inspection point. The second is to use a least squares method considering the reflection efficiency of the elastic waves at the inclined crack and the directivity characteristic of the sensor.

抄録全体を表示

In recent years, many universities have already launched small satellite on-orbit. Attitude information is important for satellite operation. However, many nano-satellite communication systems used amateur radio by narrow bandwidth frequency segment. Therefore, down-link of attitude information is expected to require substantial time. In this paper, we propose establishment of the attitude estimation method by means of the received radio power. And, satellite attitude information being estimated from the fluctuation of the received power. Then, comparison of satellite data obtained from on-board sensors with ground experiment. One major problem in this approach is the effect of Earth's ionosphere. As the radio signal passes through the ionosphere, the polarization angle is rotated by the Faraday Effect. The detection of attitude by the radiation pattern has been assumed that inaccurate. However, we are get data the spin satellite of liner polarization antenna. As a result, we can estimate 0.12rad/s accuracy of angular velocity measurement. This method can be applied to the attitude detection of many small satellites.

抄録全体を表示

The robust controller has good characteristic which can deal with disturbances and model errors. Especially μ-synthesis is a design method for the tradeoff between performance and robust stability, removing conservativeness due to unstructured singular values. We called the controller obtained by this method a μ-controller. However, it has the higher order generally. From a standpoint of cost and reliability, it is preferred that its order is reduced, which causes the degradation of the closed-loop performance. In order to decrease the degradation, it is necessary to take account the closed-loop configuration in the reduction process. For it, the authors developed the block balanced realization method in H_∞ control. In this paper, we use the scaling matrix D obtained by the D-K iteration as frequency weights for the block balanced realization method. A numerical example is given to illustrate the effectiveness of our method.

抄録全体を表示

To analyze and synthesize time-delay systems with discontinuity, the framework of differential inclusion in the sense of Filippov is extended to functional differential inclusion. Based on the extension, the concept of Filippov solution is introduced for the time-delay systems with discontinuity at first, and then it is shown that both the Lyapunov stability and the LaSalle invariance principle results can be extended to such kind of systems. Moreover, by using the proposed analysis tools, a stabilization feedback design approach is proposed for a class of nonlinear time-delay systems with discontinuity. Simulation results of numerical examples are given to demonstrate the proposed control approaches.

抄録全体を表示

In recent years, researches on autonomous robots in real life have developed. Especially, moving obstacle avoidance is one of the most important tasks for robots. Reinforcement learning is a typical method of action acquisitions of autonomous mobile robots for obstacle avoidance. However, it has been indicated that reinforcement learning has various problems in unknown environment. In order to solve these problems, we propose predictive reinforcement learning for moving obstacle avoidance. In predictive reinforcement learning, our rules are not defined as a pair of actions and states like conventional reinforcement learning. The rules are defined as the transition of the states by robot action between steps. We think that proposed rules enable robots to adapt to unknown environment because these rules are independent from any environment where moving obstacles exist. The robots implemented these rules predict the next state. After this prediction, the robots reinforce its rules by comparing observed states with predicted ones and foresee collisions on obstacles. Then the robots select safer actions. In this paper, we verify the efficiency of our method in several simulations. First, the robot is trained in learning environment where moving obstacles exist. After that, we experiment to verify the ability of adaptation to unknown environments. As a result, the robot acquires moving obstacle avoidance actions.

抄録全体を表示

This paper presents a model of decision-making considering the risk assessment. The conventional evaluation in AHP is considered to be a kind of utility. When dealing with the risk, however, it is necessary to consider the probability of damage. In order to take risk into decision-making problem, we construct AHP based on expected utility. The risk is considered as a related element of criterion rather than criterion itself. The expected utility is integrated, considering that satisfaction is positive utility and damage by risk is negative utility. Then, evaluation in AHP is executed using the expected utility.

抄録全体を表示

We developed a prototype system to support telecommunication by using keywords selected by the speaker in a videoconference. In the traditional presentation style, a speaker talks and uses audiovisual materials, and the audience at remote sites looks at these materials. Unfortunately, the audience often loses concentration and attention during the talk. To overcome this problem, we investigate a keyword presentation style, in which the speaker holds keyword cards that enable the audience to see additional information. Although keyword captions were originally intended for use in video materials for learning foreign languages, they can also be used to improve the quality of distance lectures in videoconferences. Our prototype system recognizes printed keywords in a video image at a server, and transfers the data to clients as multimedia functions such as language translation, three-dimensional (3D) model visualization, and audio reproduction. The additional information is collocated to the keyword cards in the display window, thus forming a spatial relationship between them. We conducted an experiment to investigate the properties of the keyword presentation style for an audience. The results suggest the potential of the keyword presentation style for improving the audience's concentration and attention in distance lectures by providing an environment that facilitates eye contact during videoconferencing.

抄録全体を表示

In this paper, we propose a new method for Active Net model. Active Net is a lattice network model that is based on minimizing an energy to extract the region of a target object. In a conventional method, only one target is used by one net. Therefore, assuming that the image can be divided into some parts, the proposed method uses many nets and applies them to each divided image. Some simulation results are presented obtained by applying proposed method to images. In the proposed method, since the size of a net became small, processing time to one net can be shortened. Furthermore, since the distance between lattice points is short in initial net, the target at the end of an image became detectable. That is, it means that the target detection performance has improved.

抄録全体を表示

The noise is contained with the photographic image when taking a picture by CCD Image Sensor and CMOS Image Sensor. The typical noise is divided into two elements. Former is a random noise and latter is a fixed pattern noise. The random noise is generated by the variance of the sensitivity of the photodiode and the threshold of the transistor. However, it is decreased by the semiconductor processing technology and the noise reduction technology. In the development of CMOS image sensor, the random noise will become a problem. In this study, the focus is on the photon shot noise. When taking a picture in the environment of a low luminance, the photon shot noise becomes especially predominant. I proposed a method which used a smoothing filter, a median filter and the edge reinforcement with an index from luminance variance in local area of image. A proposed method and a conventional method are used and two dimensional filters were processed. PSNR (Peak Signal Noise Ratio) is used for the comparison and the evaluation. As a conclusion, it is confirmed that the proposed method can decrease the noise more to extent compared with conventional method.

抄録全体を表示

We propose a model simulating a footsteps illusion. The model is based on a time evolution equation of the extended FitzHugh-Nagumo system with two variables that correspond to the reaction and diffusion terms. Our model visualized the following two features of the footsteps illusion: (1) an apparent change in the speed of an object moving with a constant velocity and (2) brightness induction of the moving object having a constant gray value. Thus, the model succeeded in mimicking several characteristics of human vision.

抄録全体を表示

It is challenging to capture a highly accurate 3D shape of a target object using a projector-camera pair. Attempts have been made using the pinhole model to represent the camera and projector geometry and a cubic object to estimate calibration parameters. The projector cannot emit light stripes through the same pinhole when scanning by a rotatable mirror. In addition, it is preferable to estimate the parameters using a two-dimensional object. In this paper, the problem of calibrating a structured light rangefinder is addressed. Unlike previous work, the projection of a light stripe from the projector is expressed by the equation of a plane. The coefficients of each light stripe's equation are estimated by observing a reference plane from unknown viewpoints. Experimental results demonstrate a high degree of accuracy compared to conventional methods.

抄録全体を表示

From the electronic potential point of view, the origin of the work function is divided into two terms: the ‘bulk term’ and the ‘surface term’. The former is determined by the electronic states of the bulk materials, while the latter is derived from the distribution of electrons at the surface diffusing toward a vacuum. A methodology based on these two origins is provided to modify the work function. The methodology is composed of two strategies, mixing of two metals and surface modification. The first one uses their compositions to vary the work function, and can be classified according to the phase diagram of the two metals. The second one includes the variation of surface termination, and adsorption, accumulation or segregation of the second metal. Using an analogy of surface term in a solid-vacuum interface, the relationship between work function and band alignment is considered. Then, a methodology to alter the band alignment is described as an analogous way to modify the work function. In addition, examples of the Schottky barrier height modification using such methods are discussed.

抄録全体を表示

In Totally-Implantable Artificial Heart(TAH) system, Transcutaneous Optical Information Transmission System(TOITS) and Externally Coupled Transcutaneous Energy Transmission System(ECTETS) are effective for driving, controlling and monitoring the TAH. Using these systems, a patient needs to wear an optical coupler and a transcutaneous transformer separately, and then the patient's QOL (Quality of Life) will be deteriorated. In this paper, for improving the QOL, the energy and information transmission using an unified transcutaneous transformer was investigated, and their information transmission characteristics were evaluated.

抄録全体を表示