A quasi-buried hetero-structure (BH) quantum wire (QWR) distributed feedback (DFB) laser was realized by one time selective metalorganic chemical vapor deposition (MOCVD) on a ridge substrate with a sub-micron grating. One time selective MOCVD growth forms ridge waveguide with BH structure and a QWR array for gain guided DFB LD without additional etching or re-growth process. The threshold current is 15mA, and the threshold current density is 850A/cm². A stable single longitudinal mode was preserved until 3 I_th, after which another mode emerged at higher drive current at 813.6nm. This suggests a complex coupled DFB mode operation. Elimination of the re-growth step also enlarges the selection of material for extended wavelength and operational temperature.

View full abstract

The aim of this study is to make clear the effect of temperature on the potentials of silver-silver chloride(Ag-AgCl) electrodes in potassium chloride(KCl) and sodium chloride(NaCl) solutions for micro biosensor applications. Non-isothermal coefficients, or thermal coefficient of Ag-AgCl electrodes in the concentrations of 0.1 to 1.0[mol/kg] were experimentally measured in the range of 15 to 45[°C]. The electrode was prepared by electrolytical process in KCl or NaCl solutions. The length and diameter of the electrode are 20[mm] and 3[mm], respectively. The coefficients in KCl solutions were 0.42[mV/°C] in 0.1[mol/kg] and 0.26[mV/°C] in 1.0[mol/kg] and almost the same in NaCl solutions. Isothermal coefficients were also measured. Although the coefficients must theoretically be zero by definition for all temperatures, the difference of potentials between a pair of electrodes which is called as offset voltage appears practically. They were 3∼4[μV/°C] in 0.1[mol/kg] and 0.3[μV/°C] in 1[mol/kg] at the steady state of temperature. The values were depending on the degree of the uniformity of electrode surface.

View full abstract

The rapid dissemination of the World Wide Web raises the issue of the Web accessibility, and one of the important things is the combination of a foreground color and a background color. In our previous study, the visibility of web-safe colors on the white background was examined, and the blue used for unvisited standard link color was found high visibility in wide range of ages. Since the usage of the blue and an underline are recommended as a link, in this study, we examined high-visibility background colors to the unvisited standard link color, i.e. blue. One hundred and twenty three background colors to the blue were examined using pair comparison method, and the relationship between the visibility and the color difference was discussed on the uniform color space, CIELAB (L^*a^*b^* color space). As the result, effective background colors to the standard link color were determined on the CIE LAB, that is, L^* larger than 68, a^* smaller than 50, and b^* larger than -50 provided high visibility in wide range of ages.

View full abstract

There can be seen in many situations where office workers have to make decisions under the time restriction. For example, flight control buisiness, dealing work and so on. Such the time restriction can have the time pressure, which is considerd one of mental workload, on them. In addition, it can cause human error on imformation processing. If mental work load appropriately maintained the work level, office workers would be able to go on the work with moderate tention. The purpose of this study is to quantitatively evaluate physiological indexes, which are ECG, EEG, Facial skin thermo image by using adding tasks.

View full abstract

This paper describes calculation method of equivalent relative dielectric constant εs of VVF cable. For Power line communication (PLC), VVF cables in homes are used as 2M to 30MHz high frequency transmission media. As they are not designed and installed for such transmission media, there are many branches and impedance mismatch points that cause PLC signal reflections and standing waves. In order to estimate such standing wave generation and estimate undesirable PLC signal radio wave leakage, the length of cable and relative dielectric constant εs is important. However the insulator of VVF cable is thin and not symmetric, it is not easy to calculate εs mathematically. In this paper, equivalent εs is calculated numerically using PC soft EXCEL by dividing the electromagnetic field lines and analyzing the effect of insulator for each field line. The calculation result of VVF cable considering the limited insulator area around the conductors shows that the equivalent εs is about 13% less and transmission speed is 7% faster than the insulator is full space around the cable.

View full abstract

We are proposing “Ubiquitous Acoustic Spaces”, where each sound source can emit some address information with audio signals and make us automatically access to its related cyber space, using handheld devices such as cellphones. In order to realize this concept, we have considered three types of extraction methods, which were an acoustic modulation, an audio fingerprint, and an audio watermark technique. Then we have proposed a novel audio watermarking technique, which enables contactless asynchronous detection of embedded audio watermarks through speaker and microphone devices. However its embedding data rate was around 10 [bps], which was not sufficient for embedding generally used URL address texts.
Therefore, we have extended the embedding frequency range and proposed a duplicated embedding algorithm, which uses both previously proposed frequency division method and temporal division method together. By these improvements, possible embedding data rate could be extended to 61.5 [bps], and we could extract watermarks through public telephone networks, even from a cell phone sound source.
In this paper, we describe abstracts of our improved watermark embedding and extracting algorithms, and experimental results of watermark extraction precision on several audio signal capturing conditions.

View full abstract

This paper presents a method for gender and age estimation which is robust for facial pose changing. We propose a feature point detection method which is the Adapted Retinal Sampling Method (ARSM), and a feature extraction method. A basic concept of the ARSM is to add knowledge about the facial structure into the Retinal Sampling Method. In this method, feature points are detected based on 7 points corresponding to facial organ from face image. The reason why we used 7 points to basis of feature point detection is that facial organ is conspicuous in facial region, and it is comparatively easy to extract. As features which is robust for facial pose changing, a skin texture, a hue and a gabor jet are used for the gender and age estimation. For classification of gender and estimation of seriate age, we use a multi-layered neural network. Moreover, we examine the left-right symmetric property of the face concerning gender and age estimation by the proposed method.

View full abstract

Recently, resolution conversion methods for coded images are actively studied on the field of information compression. In traditional methods, resolution conversion is performed after decoding. However, this method requires large memory and computational costs. We propose a method to decode images and to convert resolution simultaneously for improving this. Finally, we confirm effectiveness of our proposed method by simulations.

View full abstract

We have developed an eye-gaze input system for people with severe physical disabilities, such as amyotrophic lateral sclerosis (ALS) patients. This system utilizes a personal computer and a home video camera to detect eye-gaze under natural light. The system detects both vertical and horizontal eye-gaze by simple image analysis, and does not require special image processing units or sensors. We also developed the platform for eye-gaze input based on our system.
In this paper, we propose a new web browsing system for physically disabled computer users as an application of the platform for eye-gaze input. The proposed web browsing system uses a method of direct indicator selection. The method categorizes indicators by their function. These indicators are hierarchized relations; users can select the felicitous function by switching indicators group. This system also analyzes the location of selectable object on web page, such as hyperlink, radio button, edit box, etc. This system stores the locations of these objects, in other words, the mouse cursor skips to the object of candidate input. Therefore it enables web browsing at a faster pace.

View full abstract

In this study, we are aimed for the posture estimation that considered influence to lighting changes. We proposed the algorithm that considers a bad condition by using 3-D data and 2-D animation. The decrease at the gaze forward rate caused by the psychological condition and becomes tired is a problem of safety driving. There is a technique for detecting looking away and looking down for a long time by estimation driving person's face posture from the image as this method of settlement. However, conventional technique using images in the posture estimation, that is a problem of receiving a strong influence to the lighting condition and the size of the database. In this technique, it pays attention to strong for the lighting changes 3-D data and texture information. Making the database solved these problems, accuracy average was obtained 89[%] and estimation speed was obtained 6[f/s].

View full abstract

We compare the performance of four types of controllers for an autonomous mobile robot through computer simulation. Each controller consists of different kinds of architectures of neural networks (NN). Four types of NNs are as follows: Sigmoid NN, Radial NN, pulsed NN (PNN), and PNN incorporating short term synaptic plasticity (PNNSSP). In computer simulation, parameters of NNs are adjusted using genetic algorithm. The results of computer simulation show that PNNSSP is more efficient than other typs of NNs for the controller of the autonomous mobile robot.

View full abstract

Genetic Network Programming (GNP) is an evolutionary computation which represents its solutions using graph structures. Since GNP can create quite compact programs and has an implicit memory function, it has been clarified that GNP works well especially in dynamic environments. In addition, a study on creating trading rules on stock markets using GNP with Importance Index (GNP-IMX) has been done. IMX is a new element which is a criterion for decision making. In this paper, we combined GNP-IMX with Actor-Critic (GNP-IMX&AC) and create trading rules on stock markets. Evolution-based methods evolve their programs after enough period of time because they must calculate fitness values, however reinforcement learning can change programs during the period, therefore the trading rules can be created efficiently. In the simulation, the proposed method is trained using the stock prices of 10 brands in 2002 and 2003. Then the generalization ability is tested using the stock prices in 2004. The simulation results show that the proposed method can obtain larger profits than GNP-IMX without AC and Buy&Hold.

View full abstract

In this paper, using multi-agent simulations, the effect asset inequality has on an artificial society is analyzed. It is shown that it is possible for a sustainable society to decrease in asset inequality and at the same time increase economic activity.
In sustainable societies, the asset inequality increases as the consumption tax rate is raised, and in artificial societies where the tax rate is the same, inequality increases in the society in which agents with even small a surplus undertake unselfish actions. In sustainable societies which employ both income and consumption tax, an increase in asset inequalities leads to an increase economic activity. But, in sustainable societies which levy only the income tax, this result does not necessarily hold.
These results show that if economic activity is increased in sustainable societies where the consumption tax rate is raised for the fiscal stability, an inequality expansion is an acceptable consequence. However, the sustainable society with the highest economic activity is realized when only the income tax is levied. In sustainable societies which levy only the income tax, it is possible to decrease inequality while simultaneously increasing economic activity.

View full abstract

This paper presents a systematic and efficient search method for obtaining multiple local optimal solutions based on the stability theory of nonlinear dynamical systems. The method consists of two steps. The first step is a local search step for obtaining a local optimal solution. The second step is a search step for escaping from the stability region of the obtained local optimal solution via a decomposition point. Identifying the decomposition point must be a challenge task for the method. In this paper, bifurcations and a continuation method are applied for obtaining the decomposition point. For the purpose of illustrating the proposed method, well-known test functions including high dimensional functions are applied and the numerical results show the effectiveness of the proposed method.

View full abstract

IGCC (Integrated Gasification Combined Cycle) is expected to be one of the major power generating systems in the next generation. However, there are a number of problems to be solved before IGCC is fit for commercial use. One such problem is that of establishing the operation technology of the gasifier. The operation technology of the gasifier is indispensable to achieving a good balance between efficiency and availability. However this operation technology has yet to be established. We therefore developed an operation support system that finds the optimal control parameters of the gasifier, i.e. air ratio and coal feeding ratio, that maximize a calorific value of syngas, and maintain an adequate combustor temperature.

View full abstract

Iterative optimal calculation methods have been proposed for degraded static image restoration based on the multiresolution wavelet decomposition. However, it is quite difficult to apply these methods to process moving images due to the high computation cost. In this paper, we propose an effective restoration method for degraded moving image by modeling the motion of a moving object and predicting the future object position. We verified our method by computer simulations and experiments to show that our method can achieve favorable results.

View full abstract

This paper proposes a method for image contrast enhancement by using not only the distribution of gray levels but also the spatial features included in an objective input image. A contour appears on the boundary between the two planes that have different gray levels. For image contrast enhancement, expanding the difference of the gray levels in the planes that locates on both sides of the contour is supposed to be effective. By expanding the subtraction between the gray levels in the neighboring planes, the contour becomes conspicuous by visual. Therefore, the difference between the gray levels of the pixels pair is treated as the weight for contrast enhancement in the proposed method. The resultant image by the proposed method had better contrast and less noises than the conventional methods.

View full abstract

A system, which allows computer input without keyboard, is proposed. The system utilizes display-mounted Web camera for acquisition of users face and also display-mounted lamp for illumination to users. It is found that the proposed system allows computer input almost perfectly (90%) if the distance between user and display is within 30cm and if keyboard image is displayed onto a 19 inch of computer display (5cm of key distance) with 40W of fluorescent light utilized normal illumination from both side of the display. It is also found that the proposed system allows user movement because moving picture of the user face acquires in a real time basis. The relation between allowable movement of the users' motion and success rate and the relation between signal to noise ratio as well as contrast of the acquired users' image and success rate are clarified.

View full abstract

The Elevator Group Supervisory Control Systems (EGSCS) are the control systems that systematically manage three or more elevators in order to efficiently transport the passengers in buildings. Double-deck elevators, where two elevators are connected with each other, serve passengers at two consecutive floors simultaneously. Double-deck Elevator systems (DDES) become more complex in their behavior than conventional single-deck elevator systems (SDES). Recently, Artificial Intelligence (AI) technology has been used in such complex systems. Genetic Network Programming (GNP), a graph-based evolutionary method, has been applied to EGSCS and its advantages are shown in some papers. GNP can obtain the strategy of a new hall call assignment to the optimal elevator when it performs crossover and mutation operations to judgment nodes and processing nodes. Meanwhile, Destination Floor Guidance System (DFGS) is installed in DDES, so that passengers can also input their destinations at elevator halls. In this paper, we have applied GNP to DDES and compared DFGS with normal systems. The waiting time and traveling time of DFGS are all improved because of getting more information from DFGS. The simulations showed the effectiveness of the double-deck elevators with DFGS in different building traffics.

View full abstract

In recent years, wavelet transform which gives expression in time-frequency domain has been adopted for a nonstationary process. Authors have proposed a new spectral analytical method in time-frequncy domain. In the conventional spectral analytical methods, the decomposition and the reconstruction of signals have been obtained analytically. The proposed method consists of two approaches. We call these approaches as Approach 1 and Approach 2. In this paper, we show the decomposition and reconstruction of signal based on Approach 1. First, we show that the signal is decomposed to the signal elements. They are called the decomposition waves, and all these decomposed waves are added to obtain the reconstructed wave. Next, we show that the magnitude characteristic and the phase characteristic between the original signal and the reconstructed wave. Then the conditions between the signal and the reconstructd wave are derived to realize a sufficiently approximated wave. By a numeric calculation example, we show the approximation ability by the proposed method.

View full abstract

This paper discusses a real-time imputation method for sparse floating car data (FCD.) Floating cars are effective way to collect traffic information; however, because of the limitation of the number of floating cars, there is a large amount of missing data with FCD. In an effort to address this problem, we previously proposed a new imputation method based on feature space projection. The method consists of three major processes: (i) determination of a feature space from past FCD history; (ii) feature space projection of current FCD; and (iii) estimation of missing data performed by inverse projection from the feature space. Since estimation is achieved on each feature space axis that represents the spatial correlated component of FCD, it performs an accurate imputation and enlarges information coverage area. However, correlation difference among multiple road-links sometimes causes a trade-off problem between the accuracy and the coverage. Therefore, we developed an additional function in order to filter the road-links that have low correlation with the others. The function uses spectral factorization as filtering index, which is suitable to evaluate the correlation on the multidimensional feature space. Combination use of the imputation method and the filtering function decreases maximum estimation error-rate from 0.39 to 0.24, keeping 60% coverage area against sparse FCD of 15% observations.

View full abstract