Because a series of spikes in the auditory nerves are phase-locked to the appearance of either maxima or minima, the recording of extrema plays an important role in the succeeding extraction process. Approximate twenty-percent of extrema, for instance, could be reduced in the previous study according to the forward masking, which demonstrates the short-term (< 3ms) inaudible period between envelope maxima in a narrow-bandwidth waveform. Further extraction as well as the simplified modeling of acoustic structure, however, remains unresolved, still yielding a large number of extrema with minute energy or amplitude. In this study, each amplitude envelope is approximated by a kernel function, which consists of a short series of extrema. After setting a limen in line with both subjective scores and intelligibility of speech syntheses, we deleted every kernel-function below the limen, extremea between envelope maxima, and all the minima besides. Consequently, the reduction in total has reached over seventy-five percent, so that we have proposed the modified model, in which a kernel can be reconstructed by the linkage of one term of sinusoidal curves with either linearly increased or decreased amplitude, instead of a half term of sinusoids.

View full abstract

The harmonic structure, such as speech, can be divided into two parts of envelope maxima and minima. The former carries semantic information; the latter is regarded as inaudible period according to the forward masking effect. In this study, we recalculate phase error and amplitude error of minute extrema between envelope maxima, and visualize the robust structure of amplitude envelope in the light of noise allowance, corresponding to subjective thresholds. Compared with 9% phase error and 40% amplitude error in the previous study, the allowances of both indices are quite large in the concave part of amplitude envelope, estimating at 14% and 100%, respectively. Taken together with the adaptation of firing rate in the auditory nerve fibers, we have hypothesized that the marginal parts play a physiologically indispensable role in recovery from the refractory period in the auditory nerves, allowing the modification of silence with zero-amplitude.

View full abstract

Targeting at archery athletes in universities, we analyzed the relation between their performance scores and their pre-game (pre-warm-up and just pre-match) conditions which were consisted of the blood pressures and blood rates as physiological indicators and mood as psychological indicators. As a result of the analysis, it was found that the performance scores tended to be high in those athletes who showed low blood pressures and heart rates before a warm-up and that showed high scores of mood (an activation level, stability level, and comfort level) immediately before a match. Also, we tried to assess the performance levels of the archery athletes by putting the fuzzy inference (the product-sum-gravity method) into practice on the basis of the physiological and psychological indicators such as pre-match (pre-warm-up) blood pressures, heart rates, mood, and etc. As a result of this study, it was found that there tended to be a difference in the barycentric positions which were calculated by the fuzzy inference between those who had high performance rankings and those who had low performance rankings. From these study results, it was suggested that the blood pressures, heart rates and mood before a match of archery related to the performance scores and also that the assessment of the athletes could be available by putting the fuzzy inference into practice.

View full abstract

A personal authentication becomes ubiquitous because the number of Internet users is increasing. People are used to input their name and password but a password-based authentication is vulnerable because it is easy to leak a password. For this reason, people will need an authentication method which ensures security of personal information. In this article, we propose a gesture authentication method, which is one of biometrics authentication based upon some behavioral traits. It uses self-organizing maps (SOM) to authenticate users because it can learn each gesture traits and secular change. We experimented with a game controller which mounts acceleration sensors and evaluated an availability of multi-maps gesture authentication. As the result, our method obtains over 90% accuracy but map size or the number of subjects in a map seems to be ill-fitted. We will verify the accuracy of authentication under other conditions of varied sizes of maps and varied numbers of subjects.

View full abstract

Wisdom is considered as a kind of judgment based on the retrieved and extracted knowledge from miscellaneous information, and wisdom promotes the corresponding action. If we can get the wisdom from other person widely, we can do the effective action to achieve our purposes. When knowledge and information based on one's experience are understood, it becomes wisdom. So to get the wisdom it is the first step to collect a lot of knowledge which is composed of a set of low-level information, for instance, the component attributes of knowledge are situation, purpose, time, action and result of expectation. So we call them for a knowledge capsule, and we try to evaluate the knowledge capsule on the effect and cost. This paper describes the study to collect a lot of low level knowledge for Project management by the Prototype system using WEB technology. This system has the input form and has the function to search information. By the way, Team work is necessary to accomplish the project. So, it is the problem how to collect the knowledge capsule and how to use them to accomplish the project by team work in future.

View full abstract

For tracking multiple moving objects by using the particle filter, if each particle stores the states of all objects, it is necessary to use exponentially many number of particles. In this paper we propose a method in which motion samples and a window are used for the prediction phase of the particle filter. This method is useful for the tracking of multiple objects which repeat a small number of motion patterns. The proposed method enables a more exact prediction and causes the reduction of the number of particles. We show the effectiveness of the proposed method by experiments in which 5 balls in the same shape and color are juggled and successfully traced by using even a low frame rate video camera.

View full abstract

In recent years, dielectrophoresis has attracted much interest because of its effective method to achieve the separation, detection and characterization of particles such as biological cells and virus. Diectrophoresis (DEP) is a movement of particles in a given medium by a force, which is generated in non-uniform electric fields. In this paper, we measured the positive dielectrophoretic force (DEP force) in the Ceiling Creek Gap electrode. The measuring method is employed the inclined gravity, which is a component of gravitational force along the inclined floor. It is expected that the Ceiling Greek Gap electrode could identify Biological cells in blood by difference in dielectric property.

View full abstract

This paper discusses the analysis of information skills for students based on a Visual Analog Scale (VAS). The subjects were 35 undergraduate university students and 5 graduate school students and in order to analyze their information skills, they were asked to answer questionnaires on VAS and Likert Scales (LS). The results showed that there exists a high correlation between VAS and LS. The experiment data were analyzed by analysis of variance method, and the results showed the significant difference between 1st grade students and 3rd grade students. By the way, it was not clear that the different of information skills between undergraduate university students and graduate school students. VAS was shown to be an effective tool for analyzing the information skills of students.

View full abstract

It is important to reveal the ion truck structure of high energy charged particles for the field of engineering and medicine. We have developed a 3D Monte Carlo code and selected an inorganic scintillator CsI as target. So we calculate the elastic scattering of ion and electron in terms of the theory of relativity to make scattered electron angle calculation in 3D Monte Carlo code. We calculated the spatial distribution of energy by secondary electrons when H and He ions are incident. We compare the result from our 3D Monte Carlo code with a macroscopic model. We confirm large fluctuation of energy deposition from our calculation.

View full abstract

A temporal subtraction image can be used for enhancing interval changes such as shapes of new lesions and the interval changes in existing abnormalities on medical images by removing most of normal structures. There is no technique for image registration based on 3-D thoracic MDCT images until now. In this paper, we propose a new registration method for detecting the temporal changes from MDCT images based on analyzing the vector flow. The new method was applied to 5 clinical cases of MDCT images and satisfactory experimental results are obtained. The registration accuracy was improved obtained experimental results by using vector flow.

View full abstract

In this research we study a supporting method for making recipe for a measure against metabolic syndrome. We use existing recipes in order to make new recipes for a measure against metabolic syndrome. Some food materials which have less calories, are used in new recipes for a measure against metabolic syndrome instead of original food materials in existing recipes. We make a food material thesaurus. The food material thesaurus is used in the supporting method for making recipe for a measure against metabolic syndrome. We show effectiveness of the proposed method by some experiments.

View full abstract

We draw a fuzzy graph automatically using genetic algorithm (GA). As a crossover in GA, the expanded inversion is used. At first, we give the drawing rules for a fuzzy graph drawing. Then, the nodes are randomly placed at the frames in the lattice. And then the nodes are rearranged using GA so that the graph satisfies the drawing rules. By using this type of inversion the graph with one kind of fuzzy edge length with the expand display region and with the correspondence between the degree of relation of nodes and the fuzzy edge lengths are drawn, the availability of the fuzzy graphs are an estimate. Also, we show that the drawing method is useful for various purposes in sociogram analysis and describe that the method is useful for fuzzy graph drawing problems.

View full abstract

The inversion crossover in the Genetic Algorithm (GA) is useful for solving the twist of the gene and Hybrid GA combined with the inversion is effective for the TSP. When the inversion is used in a simple GA (sGA), it cannot show a high performance except the special case such as having a long size chromosome. Thus the inversion has been hardly used. Based on such a background, this paper tries to improve the convergence characteristics of an inversion. At first, we propose the inversion which the gene are exchanged according to the result of the gene evaluation. The next, we propose the multi-individuals inversion and then show that this method is effective to improve both of the rate and time of convergence. In the last, through the automatic fuzzy graph drawing, we show that the methods of the inversion proposed in this paper indicate higher performances in convergence characteristics than the usual inversion, Maximal Preservative Crossover (MPX) and Improved Edge Recombination Crossover (IERX).

View full abstract

In recent years, a treatment of implant which is technique to embed an artificial dental root in a direct jawbone as a treatment to supplement the natural tooth has been developed. It is necessary to perform a simulation to embed an implant to the correct position without injuring a blood vessel and the inferior nervous system before an operation. However, the burdens to a doctor increase to perform many manual operations for a series of simulation because of long computational time to simulate the implant. Therefore, in this paper, we propose a segmentation technique on inferior nerve canal region from dental CT images by using Hessian matrix. We applied our new method to CT images and satisfactory results are obtained. Some experimental results are shown with discussions.

View full abstract