Faculty Profiles - YAMAZAKI Tatsuya

Papers - YAMAZAKI Tatsuya

Division display 21 - 40 of about 103 ／ All the affair displays >>

Study on Automatic LED Monitoring System for Data Center Devices.

Misheel Enkhbaatar, Tatsuya Yamazaki

IEEE 15th International Conference on Computer Sciences and Information Technologies 1 340 - 343 2020

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

Monitoring is an essential process of data center operation. There are some common methods to detect data center device corruption or any type of intrusions in the operation. The best practices of monitoring data center are ensuring physical features of data center environment such as room temperature, humidity, and usage of the central processing unit (CPU) and storage of the machines which is used for hosting user services. Due to exceeding varieties of physical feature combination in data center environment, data center managing organizations cannot fully monitor it and experience unexpected conditions from time to time. Many organizations who manage data center also apply visual monitoring of Light Emission Diode (LED) indicators of devices with human interaction which takes large amount of cost and labor to reduce service interruption occurrences. In this paper, we present the concept and experimental results of an automatic monitoring system for data center devices. The concept of the automatic monitoring system is based on monitoring frequency of flickering LEDs of the device method. To evaluate performance of the automatic monitoring, we applied it to video recording of flickering LED of server device. During the implementation stage, we successfully located LED position from frame images and visualized series duration time of flickering LEDs. The experimental results present evaluation of the proposed method.

DOI： 10.1109/CSIT49958.2020.9322008

Other Link： https://dblp.uni-trier.de/db/conf/csit/csit2020-1.html#EnkhbaatarY20

First impression factor analysis through facial video evaluation experiments

Tatsuya Yamazaki

Facial Expressions: Recognition Technologies and Analysis 169 - 184 2019.5

Publishing type：Part of collection (book)

To construct a good interpersonal relationship, people generally try to make a favorable first impression on other persons. In the process of first impression construction, faces relatively rule impression formation. Since what causes the favorable first impression is still unclear, some persons are good at constructing or improving their impression and others are not. We study to find out the cause of impression improvement by conducting subjective evaluation experiments using facial video images. Two different kinds of videos are used: one is that of uttering the fixed words in a natural way (the natural video) and the other is that of reading the fixed story in such a way as to give a good impression to someone (the impressed video). A person appearing in the video is called a subject and the person who evaluates the subject is called an evaluator. For evaluation, the evaluators score the impressed videos by Semantic Differential method. Using Principal Component Analysis (PCA), two main factors on impression are extracted as "steadiness" and "forcefulness." Moreover, using two kinds of videos, the impression improvement is evaluated by the Double Stimulus Continuous Quality Scale method and the subjects are divided into two groups, that is the high and low improvement groups. By the two-way analysis of variance for the PCA factors, it is found that evaluation of "steadiness" lowers for the high improvement group. In addition, it is revealed that both genders (male and female) equally evaluate impression improvement because there is no significant difference.

Preface

Tatsuya Yamazaki

ACM International Conference Proceeding Series V 2019.1

Water management based on precision agriculture for pears

Tatsuya Yamazaki, Kazuya Miyakawa

Pears: Cultivars, Production and Harvesting 205 - 223 2019.1

Publishing type：Part of collection (book)

Water management is one of the key practices in fruit production worldwide. It is not so straightforward when to and how to control water in agriculture fields. Farmers rely on their experience and intuition and visit their farm day after day to check whether the products need to be irrigated or not. Recently, Precision Agriculture including sensing technologies is prevailed to support their decision in water management. In this paper, targeting at pears, effects of Precision Agriculture are verified for water management, which consists of irrigation and humidity measurement. For irrigation, a machine learning scheme is introduced to support farmers’ decision on irrigation. A learning mechanism is proposed where soil moisture sensing data are used. For humidity, a small sensor is used to carry out the measurement of temperature and humidity in fruit bags for pears. As the result, although there is little difference of the temperature change regardless fruit bag types, difference of the humidity change is observed for each fruit bag. Finally, pear fruit outlook quality is evaluated manually and it is related with the inner-bag factors by a statistical model.

Case studies on biometric application for quality-of-experience evaluation in communication

Tatsuya Yamazaki

Advances in Biometrics: Modern Methods and Implementation Strategies 227 - 247 2019.1

Publishing type：Part of collection (book)

With the popularization of portable terminals, the frequency of application use in the mobile environment increases, and the demand for QoS (Quality of Service) is also increasing. In the mobile environment, the communication fluctuation is large, and the quality requirement for an application that frequently transmits user’s operation to the server side, e.g., a game, becomes severe. QoE (Quality of Experience) has become one of the important topics as a mobile communication ultimate metrics in the academic field as well as in the business field. In this study, we focus on Internet service quality evaluation by biometrics. We aim to analyze the relationship between the quality of the communication service and the biometric signal in a mobile role-playing game situation as a case study. Therefore, we originally build an experimental environment to control delay as a QoS parameter, collect EEG (electroencephalography) and subjective QoE assessment data during game operation, and analyze correlation between the QoE evaluation and the EEG data of gaming. We verify the relationship between the biological signal and QoE in this study.

DOI： 10.1007/978-3-030-30436-2_11

A Narrow-Area Specific Messaging System for Tsunami Evacuation.

Naoki Kobayashi, Tatsuya Yamazaki

IEEE 8th Global Conference on Consumer Electronics(GCCE) 1032 - 1033 2019

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

For Tsunami evacuation, specific evacuation information needs to be transmitted to each evacuee, for example, the nearest evacuation facility. At present, such an information transmitting system for narrow specific areas has not been organized well. In this paper, we propose a narrow-area specific messaging system where taxis play a role of information transmitters. Since taxis have a mobility property, their allocation pattern can be changed within the target area. Effectiveness of information transmission according to the taxi allocation pattern is evaluated through Multi-Agent System computer simulation. As the experimental results, it is found that decentralized and unbiased taxi allocation can accelerate evacuation completion.

DOI： 10.1109/GCCE46687.2019.9015452

Other Link： https://dblp.uni-trier.de/db/conf/gcce/gcce2019.html#KobayashiY19

Load Profile-Based Coordination of Appliances in a Smart Home.

Rui Teng, Tatsuya Yamazaki

IEEE Trans. Consumer Electron. 65 ( 1 ) 38 - 46 2019

Publishing type：Research paper (scientific journal)

Fine-scale load profile monitoring in smart home systems provides large sources of data for the energy consumption patterns of various home appliances. New opportunities for home energy management with high-level granularity can rise due to the increasing availability of load profile data. Such energy management is especially important for hybrid energy sources, which have limited power capacity for renewable energy. This paper introduces a home energy system that collects load profiles and supports event reservation with the proactive coordination of appliances. We address the problem of profile-based coordination among home appliances, which may operate concurrently. The coordination efficiency is measured by the total energy cost, which is a function of energy prices and the energy consumption at each energy source. We propose a profile-matched time-shift approach in order to minimize the total energy cost. The proposed approach identifies a set of start times of multiple home appliances according to the power consumption prices, delay tolerances, and energy sources capacities. The proposed method is evaluated by a comparison with other methods, such as random shifting. The results show that the profile-matched shifting of home appliances yields the lowest cost compared the costs of with other approaches.

DOI： 10.1109/TCE.2018.2885874

Information Design of Evaluation Application and Process for Japanese Sake.

Yuki Hashimoto, Tatsuya Yamazaki

14th IEEE International Conference on Computer Sciences and Information Technologies 2 1 - 4 2019

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

With the progress of sensor technologies, quantitative numerical data are available for food and beverage producers. On the other hand, in most cases, consumers evaluate the food and the beverage intuitively and subjectively. It is necessary to clarify how the numerical data correlate with the mass consumer subjective evaluation. Focusing on Japanese Sake, we develop an application to collect the consumer evaluation as numerical data. Also, we propose an evaluation process with the developed application to carry out a tasting experiment. A Japanese Sake tasting experiment has been conducted by the developed application and the proposed process with 22 subjects. The questionnaire opinions from the subjects reveals that VAS (Visual Analog Scales) is appropriate for the input interface of evaluation data in the application, because evaluation of VAS is 3.6 out of 5. Also, it is found that detailed description on the evaluation items of VAS is needed for some items with ambiguous definition.

DOI： 10.1109/STC-CSIT.2019.8929764

Other Link： https://dblp.uni-trier.de/db/conf/csit/csit2019-2.html#HashimotoY19

Alert Application to Texting While Walking of the Improving Momentary User Experience and Increasing the user's intention to use the proposed that. Reviewed

Eguchi Masato, Miyoshi Takumi, Niitsu Yoshihiro, Yamazaki Tatsuya, Ohno Takehiko

The Transactions of Human Interface Society 20 ( 2 ) 243 - 254 2018

Language：Japanese Publishing type：Research paper (scientific journal) Publisher：ヒューマンインタフェース学会

Walking-while-texting on mobile phones has become a large social problem because conflict accidents with other people frequently occur in the traffic areas such as the station yards. Applications to prevent the walking-while-texting have already been launched, but they do not seem to be used commonly. In this paper, we have considered two new view points, an internal factor of screen watching and an external factor of environmental risk conditions in referring to the knowledge obtained by the group interview. We have developed a new system to timely alert to the walking-while-texting user by using a smartphone application and environmental measurement sensors. As a result of comparison of the proposed system to the application officially provided by NTT docomo, we successfully achieved the application efficiency, the effect of prevention of walking-while-texting, improving momentary user experience and raising intention to use.

Study of relationship between physiological index and quality of experience for video streaming service Reviewed

Yamazaki Tatsuya, Yamamoto Ryohei, Iijima Atsuhiko

IEICE COMMUNICATIONS EXPRESS 7 ( 6 ) 218 - 223 2018

Introduction of Following Behavior and Proposal of Reductive Models for Delay Risk in Urban Evacuation Simulation Reviewed

Tamai Hiroyuki, Yamazaki Tatsuya, Owada Yasunori, Sato Goshi, Karasawa Naoyuki

Transaction of the Japan Society for Simulation Technology 10 ( 1 ) 17 - 24 2018

Language：Japanese Publishing type：Research paper (scientific journal) Publisher：日本シミュレーション学会

&emsp;Multi-agent systems are often used in evacuation simulation. The agent needs to be modeled to include psychological factors of evacuees. The following behavior, which is one of the major psychological factors, is defined as the behavior to follow other persons when he/she does not know where to evacuate. While the following behavior is known as the risk of inducing wrongly directed evacuation, it may expedite the evacuation by chaining a group of people to a right direction. The effects of following behavior are, however, almost unknown at the city level.<br> &emsp;In this paper, the agent model with the following behavior is introduced and evaluated its property at the city level. Also, two models of &ldquo;path memory&rdquo; and &ldquo;follow-up promotion behavior&rdquo; are proposed as countermeasures against the property of evacuation delaying caused by the following behavior.<br> &emsp;Through the evacuation simulation experiments assuming the tsunami disaster, it is shown that evacuation speed depends on the ratio of the agents who know an evacuation spot. Also, we show that &ldquo;path memory&rdquo; accelerates the evacuation behavior when few surrounding agents have knowledge on the evacuation spots as well as &ldquo;follow-up promotion behavior&rdquo; shows the same tendency when several surrounding agents have the same knowledge.<br>

Agent Walking Model Construction in Urban Disaster Simulation.

Tatsuya Yamazaki, Naoki Kobayashi, Yasunori Owada, Goshi Sato

IEEE 7th Global Conference on Consumer Electronics(GCCE) 355 - 356 2018

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

Computer simulation is useful to study movement of a crowd of evacuees in a large-scale urban disaster case. Multi-Agent System is often used in the simulation and agent behaviors must reflect the psychological and physical properties of human behaviors as much as possible. In this paper, we propose an evacuating agent walking model. The walking speed depends on each person and the elderly person's walking speed can be modeled based on pseudo-elderly walk experiments. Also, based on the real evacuation data at the time of the Great East Japan Earthquake, the distribution of the evacuation start time is estimated.

DOI： 10.1109/GCCE.2018.8574780

Other Link： https://dblp.uni-trier.de/db/conf/gcce/gcce2018.html#YamazakiKOS18

Urban disaster simulation incorporating human psychological models in evacuation behaviors Reviewed

Tatsuya Yamazaki, Hiroyuki Tamai, Yasunori Owada, Kiyohiko Hattori, Shin’ichi Taira, Kiyoshi Hamaguchi

IFIP Advances in Information and Communication Technology 501 20 - 30 2017

Language：English Publishing type：Research paper (international conference proceedings) Publisher：Springer New York LLC

Building evacuation simulation provides us with various knowledge and suggestion before a real disaster happens. To date, however, evacuees were often modeled as homogeneous without individual motivation in a large-scale urban simulation model, which is rather different from real human behavior. In this paper, an evacuation simulation model with human psychological models is developed for urban disaster situation. Three psychological models are actually incorporated: normalcy bias, emotional contagion bias, and sympathy behavior bias. Normalcy bias is the initial evacuation delay caused by a belief that abnormal events rarely happen. Emotional contagion is the effect of one person’s emotional state on the emotional state of people around him/her both explicitly and implicitly. Simulated experimental results show that the proposed model provides accurate evacuation behaviors than the normal behavior model without psychological consideration.

DOI： 10.1007/978-3-319-68486-4_3

Soil Moisture Sensing Experiments for Water Management in Pear Fields.

Tatsuya Yamazaki, Kazuya Miyakawa

Proceedings of the 6th International Conference on Informatics, Environment, Energy and Applications(IEEA) Part F128273 56 - 59 2017

Publishing type：Research paper (international conference proceedings) Publisher：ACM

Water management is one of the most important tasks in agriculture. However, the farmers rely on their intuition and experience to determine when and how much water should be provided. Sensing technologies have been highly developed and several test works were conducted. There are two types of sensors to measure soil moisture: soil moisture content and soil moisture tension sensors. The usage of these sensors can be different from each agricultural product and there is no validation test for pear cultivation. In this paper, we describe the validation experiment of these two types of sensors which are conducted in pear fields. After measuring the soil moisture during one-season pear cultivation, we have clarified the difference of two types of sensors . The soil moisture tension sensor can quantify the waterretaining power of soil, while the soil moisture content sensor can detect detailed fluctuation of the water content ratio in soil.

DOI： 10.1145/3070617.3070633

Other Link： https://dblp.uni-trier.de/db/conf/ieea/ieea2017.html#YamazakiM17

Matching of Energy Provisions in Multihop Wireless Infra-Structures Reviewed

Rui Teng, Huan-Bang Li, Ryu Miura, Tatsuya Yamazaki, Peter Davis

ENERGIES 9 ( 1 ) 2016.1

Language：English Publishing type：Research paper (scientific journal) Publisher：MDPI AG

Recently there have been large advances in energy technologies for battery-operated systems, including green energy resources and high capacity batteries. The effective use of battery energy resources in wireless infrastructure networks to improve the versatility and reliability of wireless communications is an important issue. Emerging applications of smart cities, Internet of Things (IoT), and emergency responses highly rely on the basic communication network infrastructures that enable ubiquitous network connections. However, energy consumption by nodes in a wireless infrastructure network depends on the transmissions of other nodes in the network. Considering this inter-dependence is necessary to achieve efficient provision of energy in wireless networks. This paper studies the issue of energy provision for wireless relay nodes in Wireless Multihop Infrastructures (WMI) assuming constraints on the total energy provision. We introduce a scheme of Energy Provision Matching (Matching-EP) for WMI which optimizes energy provision based on matching of energy provision with estimates of differentiated position-dependent energy consumption by wireless nodes distributed in the network. The evaluation results show that Matching-EP with 4%-34% improvement in energy matching degree enables 10%-40% improvement of the network lifetime, and 5%-40% improvement of packet delivery compared with conventional WMI networks.

DOI： 10.3390/en9010045

Congestion Field Detection for Service Quality Improvement Using Kernel Density Estimation Reviewed

Yuki Shitara, Tatsuya Yamazaki, Takumi Miyoshi, Kyoko Yamori

2016 18TH ASIA-PACIFIC NETWORK OPERATIONS AND MANAGEMENT SYMPOSIUM (APNOMS) 1 - 4 2016

Language：English Publishing type：Research paper (international conference proceedings) Publisher：IEEE

Owing to proliferation of smart phones, communication services such as a video streaming service are common in a mobile situation. For these services, quality evaluation and communication control based on Quality of Experience (QoE), which is the degree of a user's subjective satisfaction, is very important because the final goal of delivering high-quality service is improving user's satisfaction. QoE tends to be affected by several factors including Quality of Service (QoS). Therefore, collecting QoS data from the users who use the mobile application has become one of promising schemes to meet the QoE, which is called the crowd sourcing data. The crowd sourcing data are, however, apt to be affected by sensing errors and low accuracy. In this study, we propose to estimate some densities just from the records of application use and its location information to remove the sensing errors and low accuracy. The Kernel density estimator is used to derive a continuous density function from discrete distributed sample data such as the collected QoS data. From the viewpoint of QoS, it is important to extract high-density fields of the application use to find out QoS degradation. After estimating the Kernel density estimator, we determine the borderline between the high-density field and the other field by using the reference value that can be determined from the observed data. Simulated experiments verify effectiveness of the proposed method.

DOI： 10.1109/APNOMS.2016.7737246

Interactive Facial Expression Reader and Extension to First Impression Improver Reviewed

Tatsuya Yamazaki, Ken'ichi Maehara, Koichiro Enomoto

PERCEPTION AND MACHINE INTELLIGENCE, 2015 26-27-February-2015 183 - 188 2015

Language：English Publishing type：Research paper (international conference proceedings) Publisher：ASSOC COMPUTING MACHINERY

Facial expression plays an important role as non-verbal information in dialogue communication. People complement verbal information with non-verbal information to hold a smooth conversation. To date, collaboration between information engineering and psychology has enabled development of human facial expression recognition technology. Most of the related studies, however, have ended up with a laboratory-based demonstration. In this paper, focusing on the facial expression database, we overview the facial expression recognition technology. Then we present a simple application using the current status of the facial expression recognition technology. Finally we show prospect of the technology with a more realistic application.

DOI： 10.1145/2708463.2709059

Other Link： http://dblp.uni-trier.de/db/conf/permin/permin2015.html#conf/permin/YamazakiME15

Verification of View-based Self-localization of Autonomous Outdoor Mobile Robots Reviewed

Tatsuya Yamazaki, Shinichi Yuta, Tadahiro Hasegawa

2015 54TH ANNUAL CONFERENCE OF THE SOCIETY OF INSTRUMENT AND CONTROL ENGINEERS OF JAPAN (SICE) 1140 - 1143 2015

Language：English Publishing type：Research paper (international conference proceedings) Publisher：IEEE

We reported on the self-localization of autonomous outdoor mobile robots based on detecting the edge of pavement in camera images. The detecting technique of utilizing the difference in color between the edge and pavement was proposed and verified, to realize edge detection that is not affected by the sunshine condition of pavement in the captured image. The experimental results showed that the self-localization based on the edge of pavement in camera images is useful for outdoor mobile robots.

DOI： 10.1109/SICE.2015.7285572

Temperature and Humidity Measurement inside Pear (Pyrus Communis) Bags during Preharvest Bagging Reviewed

Tatsuya Yamazaki, Tomohiko Sugiyama

2015 54TH ANNUAL CONFERENCE OF THE SOCIETY OF INSTRUMENT AND CONTROL ENGINEERS OF JAPAN (SICE) 1199 - 1202 2015

Language：English Publishing type：Research paper (international conference proceedings) Publisher：IEEE

In order to protect fruits, preharvest bagging is usually used for fruit growth including apples, peaches, grapes and pears. Characteristic and performance of the fruit bag used for the preharvest bagging have an important role to improve quality of the fruits. Effects of preharvest bagging have been studied by surveying postharvested fruits. However, studies on the inner-bag environment including temperature and humidity have been hardly studied during fruit growing so far. In this paper, it is proposed to measure temperature and humidity inside different fruit bags to clarify difference of the bags. Targeting at pears 'Le Lectier', a small sensing logger is used to carry out the measurement for three different types of bags. Based on 130 days continuous measurement, the difference of the bags are gured and visualized.

DOI： 10.1109/SICE.2015.7285426

Resource Allocation Method Based on QoE for Multiple User Types Reviewed

Tatsuya Yamazaki, Takumi Miyoshi

7th International Conference on Communication Theory, Reliability, and Quality of Service (CTRQ2014) 13 - 18 2014.2

Language：English Publishing type：Research paper (international conference proceedings)

Both of Quality of Service (QoS) and Quality of Experience (QoE) are defined to specify the degree of service quality. In some sense, QoE includes subjective evaluation from the users as an extension of QoS. Therefore feedback from QoE to QoS control might realize user-oriented resource allocation in network services. Utility functions are sometimes used to assign a bridging role between QoS and QoE. Although the user characteristic has variety, a single utility function was used in previous studies in most cases. Moreover the QoS control, i.e. the resource allocation, by making use of the utility functions are hardly studied yet. In this paper, multiple user types which have respective utility functions are considered. Respective utility functions are acquired from real experiments. Then a resource allocation method is proposed to reflect each user type satisfaction based on the utility functions. The simple case is studied and the resource allocation method is derived analytically.

　 PREV - NEXT 　

1
2
3
4
5
6