In marine aquaculture, fish populations constantly decrease throughout the cultivation period because of mortality and escape. Current production management systems provide limited opportunities to count the cultured fish, making it difficult to estimate accurately the fish population in the cage. To overcome this problem, an automatic fish identifying method based on particle tracking velocimetry (PTV) flow visualization technology is proposed in this paper. The proposed method utilizes an image processing unit that extracts individual fish from the acquired image and a motion analysis unit that calculates the motion vector for each individual. Thus, the accuracy of the extraction results in the image processing unit affects the system's counting results. To validate the efficiency and robustness of the image extraction performed by the image processing unit, individuals were extracted from images using the open-source image deep learning semantic segmentation method (SegNet), which is able to distinguish between the background and foreground in the images via analysis at the pixel level. SegNet is able to improve the image discrimination performance by multiplying the learning paths, and the robustness of the detection results can be ensured by changing the layer structure according to the detection target. Accordingly, the use of SegNet was evaluated in terms of the number of layers and images in the training set. The results of this study indicate that the application of SegNet with PTV technology represents a promising method for the automatic identifying and behavioral tracking of fish in an aquaculture net cage.

In the course of study for information of high school in Japan, the contents include “understand the structure and components of the network”. However, there is little experiential learning of router or L2 switch which are real component of the network, because it is difficult to prepare as learning materials. Therefore in this study, we are developing the learning materials to enable students to practice router or L2 switch using the Raspberry Pi collectively using OpenFlow.

We have proposed a fish-farm monitoring system. In our system, the transmission range of acoustic waves from sensors attached to the undersides of the fish is not omnidirectional because of obstruction from the bodies of the fish. In addition, energy-efficient control is highly important in our system to avoid the need to replace the batteries. In this letter, we propose a data-gathering method for fish-farm monitoring without the use of control packets so that energy-efficient control is possible. Instead, our method uses the transmission-range volume as calculated from the location of the sensor node to determine the timing of packet transmission. Through simulation evaluations, we show that the data-gathering performance of our proposed method is better than that of comparative methods.

It is important for beginning network engineering students to frequently practice setting up fundamental IP networks according to predetermined procedures. We have been developing NetPowerLab, a hands-on IP network practice system that uses virtual machines to provide an IP network learning environment. In this paper, we report the new development of IP network practice system with software agents. Software agents operate as a teacher or a cooperative student. The new system has three mode. The tutorial mode is used by beginner students with software agent operating as teacher. The practice mode is used by intermediate students with software agent operating as intermediate cooperative student. The troubleshooting mode is used by advanced students with software agent operating as advanced cooperative student. We describe outline of the system and software agent for IP network practice in this paper.

In this paper, we develop a system to support learning of OpenFlow by visually associating controller configuration information and logical topology. In our system, the constructed topology is visually shown on a touchscreen display and a learner can easily confirm the topology. In addition, a learner can configure the controller by visual and simple operation using the touchscreen display. Therefore, a learner can learn the behavior of the OpenFlow even if the learner is a beginner. Through experimental evaluation, we confirmed that our system can be applied to an OpenFlow network consists of dozens of OpenFlow switches.

As a method for supporting fish farming, this paper presents a video scene detection when farmed fry start instantaneously in a tank due to environmental stimuli. Although some environmental stimuli such as sound noises or lighting startle the fry and the stimuli bring about the instantaneous response, actual situations around the tanks in which the stimuli occur are unclear in detail. From the fact the fry often die due to crashes to the tank's wall and between the fry by the response, a monitoring system for the fry and situation around the pool could find causes of the stimuli, and it could result in decrease of the death number of the fry. In this research, the fry which swim in a tank are monitored by a video cam- era and the video scenes at the response are detected by a SVM with a feature value which represents fry's acceleration using sequential frames of the moving image. Preparing the moving images which include scenes of the response by fish in a tank, performances of the proposed method were examined. From experimental results, accuracy ratios of the recall and the precision for the scene detection have shown more than 80% on average and 100% under normal illuminances (108.5 lux on average), respectively.

One of the promising sensor network application is for the primary industry and we have proposed a fish farm monitoring system. In our system, sensor data monitored at a fish-mounted sensor node is transmitted to a sink node by using acoustic communication. Since we assume that a sensor node is attached at an abdominal cavity region of fish, the transmission range of acoustic wave is not omni-directional due to obstruction of fish body. In addition, energy efficient control is highly important in our system since it is difficult to replace battery of sensor node. In this paper, we propose a data gathering method taking into account energy efficiency and directional transmission range. In our method, we use the volume of transmission range to determine the timing of packet transmission. Our method does not require control packets for determining its transmission timing. Through simulation evaluations, we show that the data gathering performance of our proposed method is higher than that of comparative methods.

In network education, there are several benefits for conducting cooperative practice among a variety of learning levels of learners. We have proposed a hands-on IP network practice system by constructing a virtual network on a cloud server. In our past study, we introduced a pseudo cooperative practice function to our system so that cooperative practice with a companion agent can be accomplished even if there is no other learner for constructing a group. In this paper, we extend the pseudo cooperative practice function to deal with a variety of learning levels of learners. In our system, the companion agent changes its role automatically depending on the learner's learning level. Through questionnaire investigation, we show the effectiveness of pseudo cooperative practice.

OpenFlow network technologies has attracted attention along with popularization of cloud environment and server virtualization since it enables flexible network configuration. It is expected that existing networks of many organizations will migrate to OpenFlow networks in the future. Since it takes a huge cost to replace all existing network devices to OpenFlow switches, many organizations consider to employ overlay OpenFlow networks where virtual OpenFlow devices are introduced at endpoint servers. In overlay OpenFlow networks, the cost for migration increases according to the number of servers. In this paper, we propose a system for supporting migration to an overlay OpenFlow network. Our proposed system acquires settings of servers in the existing network endpoints and reflects them to the overlay OpenFlow network. Through experimental evaluations, we confirm the behavior of the configuration acquisition function of our proposed system.

We have developed a screen sharing system to share contents between two persons, and applicable to one-to-one remote teaching. The screen of a PC at one side is shared with the screen of a tablet at the other side through the network to convey instructions from an operator to a collaborator. This system makes it possible to arbitrarily select a part of the screen of the PC at the operator side. The selected screen is presented in the tablet at the collaborator side. The collaborator can adjust the scale of contents, and capture the screen. By analyzing such operations, the system can understand and record which parts of the contents the collaborator paid attention to. In addition, the camera of the tablet can be used as a simple scanner to digitize paper documents easily. Further, characters and symbols drawn with a finger or a pen on the tablet screen can be presented on the PC at the other side.

Studying networking technologies has become one of the most important subjects in the field of Technology of junior high schools and Information Study of senior high schools in Japan. To study these subjects, studying with practically and experimentally is strongly recommended by MEXT (Ministry of Education, Culture, Sports, Science and Technologies in Japan). However, there is little equipment for these studies. We have been developing new educational equipment for experimental studying of networking technologies based on concept of physical visualizations and physical direct manipulations. The equipment can visualize physically how network works using LED. Also the equipment provide physical direct manipulation using dials and buttons.

Software-Defined Networking (SDN) has attracted attention along with the popularization of cloud environment and server virtualization. In SDN, the control plane and the data plane are decoupled so that the logical topology and routing control can be configured dynamically depending on network conditions. To obtain network conditions precisely, a network monitoring mechanism is necessary. In this paper, we focus on OpenFlow which is a core technology to realize SDN. We propose, design, implement, and evaluate a network monitoring system for OpenFlow networks. Our proposed system acts as a proxy between an OpenFlow controller and OpenFlow switches. Through experimental evaluations, we confirm that our proposed system can capture packets and monitor traffic information depending on administrator's configuration. In addition, we show that our proposed system does not influence significant performance degradation to overall network performance.

Education of network security has attracted a lot of attentions due to increasing unauthorized access, lack of engineers who have network security skills, and so on. Although practice using actual computer networks is highly important for network security education, it requires costs to prepare a computer network for practice. We have proposed a network security learning system by constructing a virtual network on one PC so that low-cost, easy and safe practice can be accomplished. In this paper, we introduce an attacker agent that automatically attacks servers in a fixed pattern for supporting practical learning of defense techniques safely on a virtual network.

Drumming motion of percussion is important to play music, and it can learn efficiently by coaching from expert. However, if novice drummer cannot get coaching from expert, they practice percussion by oneself with use of book. Then, it is difficult for novice drummer to measure of practice result. We have been developing self learning system for novice drummer. Using this system, novice drummer can check percentage of a much between instruction and practice.

Aerosol retrieval work from satellite data is divided into three parts: satellite data analysis, aerosol modeling, and multiple light scattering simulations in the atmosphere modeling what is known as the radiative transfer problem. It is well known that the aerosol model has been developed using the accumulated measurements during more than ten years provided with the world-wide aerosol monitoring network (AERONET). The radiative transfer simulations incorporate the Rayleigh scattering by molecules and Mie scattering by aerosols in the atmosphere. Thus the aerosol properties are estimated by comparing satellite measurements with the numerical values of radiation simulations in the Earth-atmosphere model. It is reasonable to consider that the precise simulation of multiple light scattering processes is necessary, and requires a long computational time, especially for an optically thick atmosphere. Therefore, efficient algorithms for radiative transfer problems are indispensable to retrieve aerosols from space. It is shown here that dense aerosol events can be well simulated by a semi infinite radiative-transfer model. For an example, the air pollution event observed by Aqua/MODIS in June of 2010 over Beijing is interpreted based on the proposed aerosol models and the semi-infinite radiative transfer simulation code known as MSOS (Method of Successive Order of Scattering).

Studying networking technologies has become one of the most important subjects in the field of Technology of junior high schools and Information Study of senior high schools in Japan. To study these subjects, studying with practically and experimentally is strongly recommended by MEXT (Ministry of Education, Culture, Sports, Science and Technologies in Japan). However, there is little equipment for these studies. We have been developing equipment for experimental studying of networking technologies. Connecting each peace of equipment, students can study networking configurations, such as IP address assigning, Netmask, Routing and so on. In this paper, we will describe outline of our research and the equipment.

In this study, we developed an automatic network replication system for OpenFlow network. This system can automatically and virtually replicate OpenFlow network according to captured information. This system consists of mainly two functions: Message Capture Function (MCF), Network Emulation Function (NEF). MCF is located between OpenFlow controller and OpenFlow switch. MCF can capture message information that is exchanged between OpenFlow controller and OpenFlow switch. In addition by use of LLDP, MCF can detect link information between OpenFlow switches. Then, MCF performs topology inference by link information. NEF can replicate OpenFlow network according to captured message information and link information. Users can do a test newly developed function of OpenFlow freely and easily on just one PC by use of this system.

The purpose of this study is to give students hands-on practical learning for IP network skills. We have been developing the hands-on IP network practice system using virtual machines. Students can practice IP network skills freely at anytime on just one general-purpose PC. We have implemented this system by applying the virtual Linux environment User Mode Linux. In this research, we developed two new functions for this system. First, we developed the Automatic Scoring Function. This function compares student's configuration files and teacher's configuration files. Then, this function can show a result as a percentage of correct answer. In addition, this function can display the details of items that are incorrect. Using this function, students are able to find that their network configurations are correct or incorrect by themselves. Students are able to do self-service training and testing all in one place. Second, we developed the Assignment Making Support Function. Teachers are able to make assignments by using this function easily. To evaluate new functions, this system was experimented. The results indicated that this system had sufficient performance for introductory network courses for undergraduate students.

The quality of the delivered lecture image becomes an important factor to improve an educational effect in a real time distance-education. In this paper, we pay attention to teacher's needs for the lecture video image. We developed the camera control system using a pointing stick corresponding to the teacher's need. In this paper, we describe the proposal of camera control model, the outline of our system and the experimental use of our system.

The purpose of this study is to give students a hands-on practical learning for IP network skills in campus cloud computing. It is important for students to operate network devices repeatedly to learn network skills. However, in practical learning with real network devices, students need a number of network devices. For this reason, it is not easy for students to prepare environment to be able to practice freely at any time. Therefore, we have developed a hands-on practical learning system to learn IP network skills on general performance PC. Unlike existing network simulators, this system does not require students to learn scripting language to make scenarios. This system enables students to practice network skills freely and virtually on just one PC. We have implemented this system by applying the virtual Linux environment User Mode Linux. This system consists of virtual routers, virtual hosts, and operating functions. This system is primarily intended to practice operation of routing on virtual routers. In this research, we developed new system that was designed on client server model for web services. To evaluate, this system was experimented. The results indicated that this system had sufficient performance for introductory network courses for undergraduate students. © 2012 Crown Copyright.

多くの教育機関では，ネットワーク技術者育成のための演習が行われている.
学習者は，演習において，ネットワーク機器を繰り返し操作することで，ネットワーク構築技能を習得する．しかし，実機を使った演習では，複数のネットワーク機器が必要である.このため，学習者がいつでも自由に実習できる環境を準備することは容易ではない.そこで，本研究では1台のPC上にネットワークが構築できるシステムを開発した．本システムによって，学習者は1台のPC上に自由にかつ仮想的にネットワーク構築の演習が実施できる．このシステムを仮想Linux環境であるUser Mode Linuxにより実現した．本システムは，仮想的なルータと仮想的なホスト，およびこれらの仮想的な機器を操作する機能から構成される．本研究では主に仮想ルータへのルーティングの設定操作の演習を対象とする．システムの評価実験を情報系の大学生13名に対して行った．実験の結果，ネットワークの構築演習については

The purpose of this study is to give students a hands-on practical learning for IP network skills. It is important for students to operate network devices repeatedly to learn network skills. However, in practical learning with real network devices, students need a number of network devices. For this reason, it is not easy for students to prepare environment to be able to practice freely at any time. Therefore, we have developed a learning system to learn IP network skills on just one PC. Unlike existing network simulators, this system does not require students to learn scripting language to make scenarios. This system enables students to practice network skills freely and virtually on just one PC. We have implemented this system by applying the virtual Linux environment User Mode Linux. This system consists of virtual routers, virtual hosts, and operating functions. In this research, this system is primarily intended to practice operation of routing on virtual routers. To evaluate, this system was experimented. The results indicated that this system had sufficient performance for introductory network courses for undergraduate students.

The concentration of suspended particulate matter smaller than 2.5 mu m (PM2.5) was linearly correlated with the column aerosol optical thickness (AOT) based on simultaneous measurements at a NASA/AERONET station at Kinki University Campus, Higashi-Osaka, Japan, between March 2004 and June 2006. The correlation coefficient differed with the aerosol type, being maximal when PM2.5 values were measured 120 minutes after AOT data for a dust episode, but almost independent of the time difference between measurements for anthropogenic aerosols. The obtained results were validated using data obtained at the Higashi-Osaka and Noto sites. Our results suggest that the PM2.5 mass concentration can be estimated from the AOT, and vice versa, and hence a distribution map of PM2.5 can be produced from the satellite-derived AOT map determined from the Aqua/MODIS sensor.

Our software (the EIVS) is designed for supporting environmental education by providing tools for 1) automatic data collection through networks, 2) data analysis, and 3) graph drawing of the environmental data measured by sensor-robots located at remote areas. Our software using Squeak (the Smalltalk environment) works on the hardware OLPC-XO with Squeak Virtual Machine. Our software on the OLPC-XO will allow a larger number of children using the OLPC-XO to recognize situations and problems of our environment for the discussions and the considerations of their better solutions.

In order to analyze the urban atmospheric particles in Asia, we have undertaken simultaneous measurements of aerosols and suspended particulate matter (SPM) at a NASA/AERONET station at Kinki University Campus, Higashi-Osaka, since March in 2004. The relationship between concentration of suspended particulate matter (PM) and column aerosol optical thickness (AOT) is examined. It is found that a strong linear correlation exists between PM and AOT, and better correlation is obtained for the aerosols with the same origin, such as anthropogenic aerosols and dusts, rather than overall. The derived regression lines between PM_2.5 and AOT highlight the possibility that the PM_2.5 concentration can be estimated from the AOT, and vice versa. As a result, a distribution map of PM_2.5 has been obtained by using AOT values derived from Terra/MODIS data observed in April 2006 over East Asia.

BIX is an XML standard for the data exchange between sensed data of die information arising on the spot and database. The nutrimental components in plants are important information for the diagnosis and the cultivation control of the agricultural products. Application of mid-infrared (MIR) spectroscopic and the X-ray fluorescent (XRF) spectroscopy to the simple, non-destructive, simultaneous, and rapid monitoring of the nutrient information of fresh plants is desirable as the high potential implements. The both spectroscopic methods provide substantial potential as the quantitative tools in the fields for the chemical components relating to the plant vigor. Since it is necessary to take an accurate grasp of the quality change in time and to make a suitable cultivation management for the change, we also carried on research of "the development of BIX Image Broker System (BIX-IBS)" and "the remote monitoring of color of agricultural products for evaluating the quality". BIX has to determine a standard not only for data acquisition and exchange but also for data storage, concentration and handling. © 2008 SICE.

The image monitoring system for paddy field was constructed by using BIX-IBS for growth diagnosis of rice cultivation at the field level. The image of paddy field was calibrated by the image processing node of BIX-IBS with the color calibration function since it was influenced on the outdoor natural light. As the results, it was shown that the time for farm work can be presumed from the monitoring data of paddy field. It has been found that the relation between the leaf color and SPAD are linear in the specific growth period

The aerosol properties of urban atmospheric particles have been analyzed for radiometry data obtained using a multi-spectral photometer located at a NASA/AERONET station at Kinki University Campus, Higashi-Osaka, Japan, since 2002. The suspended particulate matter has been simultaneously measured at the same AERONET site since 2004. Our measurements and a clustering analysis reveal the aerosol types over the industrial city of Higashi-Osaka. It is shown that aerosols at Higashi-Osaka can be classified into three categories: (1) ordinary, which represents the background clear atmosphere of Higashi-Osaka, (2) anthropogenic aerosol events, and (2) dust episodes called Kosa, when large amounts of soil dust are transported to Higashi-Osaka from the Chinese mainland on westerly winds, especially in spring. We also found that the linear correlation exists between column aerosol optical thickness and PM2.5 concentration, and the correlation is better within each aerosol cluster than overall.

This research proposes an ad hoc wireless communication method that is appropriate for farm networks. We plan to apply this technology in an orchard. In this paper, we addressed farm networking requirements through an ad hoc network. We proposed a communication method that considered remaining battery charge and communication priority. In this proposed method, the wireless terminal provides service corresponding to the level of remaining battery charge. In addition, the terminal communicates by using a routing table corresponding to the level of remaining battery charge. When communication throughput decreases because of changes to the routing table, the terminal secures throughput by raising the priority of IEEE802.11e EDCA when transmitting. Priority is determined from the kind of the packet, the battery level, and the relay frequency to the destination terminal. This priority is dynamically changed according to the network situation. In this research, the QoS control function of IEEE802.11e is enhanced. A routing protocol widely used on ad hoc networks, AODV, was used in this research. The proposed method was implemented on a NS2 network simulator, and an evaluation experiment was conducted. The experimental result confirmed that the proposed method was effective.

This paper shows and discusses the implementation of new ideas we propose with the aim to overcome the visual discomfort that arises when using 3D visualization. Our proposal can be summarized as the improvement of 3D visualization by replicating the way human eyes see, in other words, the increase of human-like perception. We do so by extracting reality information from stereo image pairs and process them in order to improve the final 3D image before displaying it to the user. We describe the techniques used to extract information from stereo images as well as the techniques created to manipulate them in order to produce human-like perception with the aim of reducing visual discomfort in users of the stereo visualization system. We also describe the results of a preliminary evaluation that strongly suggests the effectiveness of these techniques. © 2006 ICASE.

Image data is widely used for the diagnosis of problems with cultivation management, cultivation recording, growth level recording and pest diagnoses etc. In this research, we developed a system, Gateway-P2P-Node, to link the BIX Image Broker System (BIX-IBS) with existing network services. BIX-IBS provides efficient exchange of a variety of agricultural product digital image data. In this research we targeted Data Storage, the system that manages Field Server images, and ImageServer, an image processing application.
We developed Gateway-P2P-Node to link these network applications and BIX-IBS. BIX-IBS users can access both Field Server's Data Storage and ImageServer through the Gateway-P2P-Node. In this paper, we experimented via the Internet between Kinki University and Niigata University. The experimental results confirmed that BIX-IBS and the existing network service were able to be linked by Gateway-P2P-Node.

In agriculture, image data is widely used for the diagnosis of cultivation management, cultivation record, and growth level and diagnoses of pests, etc. In this research, we developed a BIX Image Broker System that used P2P. The purpose of this system is an efficient exchange of agricultural products images. This system was developed for Hybrid type P2P system and this system distributes and manages the meta data and the image data. By having developed as hybrid type P2P system, the search speed of the image improves and images are efficiently exchanged. In addition, we can use the function of the certification of joining in the group.
By this function, we can exchange data safely in a specific group. Users can easily search or get images by using a user-interface on a P2P node.
In this research, we developed a prototype system and then we experimented on the Internet. From the experiment result, we showed the feasibility of the P2P system that exchanges agricultural products image.

This research proposes an XML-based standard, BIX-image, for agricultural product images and image meta data exchanged through the BIX Image Broker System.
In addition, we developed a peer to peer (P2P) message handler, BIX-P2PHandler, to exchange BIX-image data as P2P messages. The BIX Image Broker System is an information sharing system developed for efficient exchange of a variety of agriculture product digital image data using P2P technology. Images and their meta data are built into P2P message and exchanged between nodes. In the work reported in this paper, we identified key attributes of agricultural product images, and defined a mapping of those attributes to XML as BIX-image. Moreover, we developed BIX-P2PHandler to transfer BIX-image data as a message between peer systems. In addition, we confirmed that we could exchange agricultural product image information between P2P nodes by developing a trial model of BIX-IBS.

SVGを用い，PDAの画面上に各種情報を統合して提示するシステムの開発

土中水分を測定するためのセンシング技術に関する研究

The purpose of this paper is to describe a new Active Queue Management and Packet Scheduling Technique to be used by real-time Internet video applications. In this technique packets are classified into classes according to its importance to the application. Also, every packet is marked with a Time Limit to arrive at the destination and the expected Number of Hops until arrival. It is used as a measure of distance. Time Limit and Number of Hops are combined to form a relay priority for forwarding and are used to find the packet a preferential position in the router output queue. Packet transmission using this technique helps us assure Quality of Service to the operation of real-time video applications.

高精度なGPSを用いて圃場の形状を測定する方法について提案した。また、測定したデータから3次元マップを生成する方法について報告した。（英文）

This paper proposes a method of video communication for remote technical consultation on the Internet. In the proposed method, images are divided into a region of interest and a region of no interest, and then images are layered by spatially frequency and coded. Priorities are given according to regions and layers for packets. In this paper, we assume that routers transmit packets according to priority and network condition. By given top priority to the region of interest, packets that are included in the region of interest are forwarded as first priority packets. Using the proposed method, we can perform remote technical consultation by video communication on the Internet.

本論文では，狭帯域回線上での動画像通信による遠隔技術相談システムのための画像QoS設定機能を提案する．提案する機能では，一画面を注目領域と非注目領域に分け，それぞれの領域について異なるQoSを設定する．本研究では，変化の大きな画像による遠隔技術相談に対応するために，条件付画素補充方式の設定をQoSの項目とする．提案する機能により，変化の大きな画像による遠隔技術相談においても，注目領域の画像のQoSを確保した狭帯域回線上での動画像通信が可能となる．この機能を遠隔技術相談用ビデオツールに実装し，機能を評価するために実際の技術相談事例を用いた実験を行った．実験結果より，提案した機能が良好に働き，遠隔技術相談に有効であることを確認した．

本論文では，狭帯域回線上でビデオ画像通信による遠隔技術相談システムに適した動的QoS制御機能を提案する．本研究で提案する機能の特徴は，画像の受信者側からの要求と技術相談の目的に応じて，注目領域の選択範囲と画像の量子化係数を動的に変更することである．提案する機能により，遠隔技術相談におけるQoSパラメータの調整作業が不要となる．さらに，使用している帯域を使い切るように，これらの値を動的に制御することで，与えられた帯域の中で可能な限り多くの情報の転送が可能となる．この機能を遠隔技術相談用ビデオツールに実装し，3つの帯域に対して遠隔技術相談に適用する実験を行った．実験結果より，いずれの帯域においても，提案した機能が有効に働き，与えられた帯域の中で可能な限り多くの情報が転送でき，ネットワーク帯域を有効に利用できることを確認した．

The purpose of this study is to develop a dynamic QoS(Quality of Service) control function for Remote Technical Consultation System. The Remote Technical Consultation System realizes technical consultation using video communication tool through narrow band networks. The proposed function can dynamically control size and quality of video images according to user's demands and purposes. The user does not need to adjust QoS parameters by using the proposed function during performing remote technical consultations. Furthermore, the proposed function can transmit video images effectively for an assigned network by controlling QoS parameters to fit video images into the network. We developed the dynamic QoS control function and added this function to video tool for the Remote Technical Consultation System. We experimented on three different bandwidth networks using the developed function. From experimental results, we verified that the proposed function works finely on any bandwidth and transmits video images to the best on the assigned network bandwidth.

This paper proposes a function for remote technical consultation using video communication through a narrow band network. The proposed function has the following features: 1) a receiver can select a part of a video image based on a user's interest without any extra device, 2) the receiver can control a size of video image, quality of the video image, and number of color (color image or monochrome image). The proposed function enables user to transmit a high quality image with smooth motion even through a narrow band network. We experimented on remote technical consultation using proposed function. From the experimental results, we verified that the proposed function works finely and we can perform remote technical consultation with a high quality image through a narrow band network.

This paper deals with the proposal of a practical method for reducing turning motion resistance of the multi-powered wheel vehicle. In the previous paper the same kind of idea as this practical method for improving the turnability of the tracked vehicle was introduced. From that discussion it was found that the proposed method can be effectively applied for actual use of tracked vehicle and it reduced the turning motion reisitance drastically compared with the normal turning motion. Through the experiments, almost the same amount of reduction of turning motion resistance as the tracked vehicle could be observed. This proposed idea of reducing the resistance can be practically applied to commercial off-road vehicle for improving the turnability and the manoeuverability for obstacle negotiation.

立体視疲労を軽減する表示方法

フィールドサーバの応用技術の紹介

コンピュータネットワークとTCP/IP の基礎の解説、および基本的なネットワークツールや、ネットワークパケットアナライザツール、ネットワークセキュリティツールについてその詳細な利用法を紹介した。

TCP/IPプロトコルの基礎と応用およびネットワークツールの活用方法

TCP/IPプロトコルの基礎と応用およびネットワークツールの活用方法

TCP/IPプロトコルの基礎と応用およびネットワークツールの活用方法

TCP/IPプロトコルの基礎の解説

クラウド環境やサーバ仮想化の普及により，SDNが注目を集めている．SDNではコントロールプレーンとデータプレーンが分離されることにより，論理トポロジや経路制御の動的な変更が可能である．一方で，ネットワークの状況を正確に把握することが困難である．そこで本研究では，SDNの中核技術であるOpenFlowを用いたネットワーク運用管理支援システムを開発した．本システムでは，実ネットワークを仮想環境上に再現することにより，SDNと従来型のネットワークの両方をテストできる．本稿では，従来型のネットワークを仮想環境に再現する機能に関する評価方法について検討する．

本研究では，詳細画像による診断中に農作物の全体像を把握できる遠隔診断システムを開発した．本システムでは，一つのビデオストリームから詳細画像と全体画像の二つの動画像を利用できる．これにより，注目領域観察時の不自由さが軽減できる．

内部処理の可視化を可能とするソフトウェアルータを開発した．NetPowerLabは仮想マシンであるUser Mode Linux(以下，UML)を仮想的なネットワーク機器として動作させ，複数の仮想マシン同士を相互に接続することで，1台のコンピュータ上で仮想的にネットワークの構築を可能とする．さらに本ルータを使用することで，ルータ内部で動作する処理の流れを視覚的に確認できる．これにより学習者は，構築したネットワーク上のルータで実行される処理の流れについて学習できる．

本研究では，実ネットワークのテスト環境を仮想的に提供することを目的とする．さらに，SDNと従来のネットワークの混在環境にも対応する．これら二つの目的を達成することで，SDNの運用を支援できる．目的を達成するため，SDNの中核技術であるOpenFlow2)を用いて，ネットワーク運用管理支援システム(以下，本システム)を開発した．本システムでは， OpenFlowネットワークと従来のネットワークの両方を仮想環境に再現する．従来のネットワークの再現にはNETCONF(Network Configuration Protocol) 3)を使用する．これにより，OpenFlowネットワークと従来のネットワークの混在する環境であってもテストが可能となる．

ネットワーク技術の初学者は，課題であらかじめ決められた手順通りにネットワ ークを構築することで，ネットワーク機器の操作方法や設定手順を習得する．課 題を用いた演習では，学習者はpingやtracerouteなどのコマンドを用いてネット ワークが正しく動作しているかを確認する．正しく動作しない場合は，コマンド の出力結果や各機器の設定から誤っている箇所を特定し，修正する．しかし，初 学者にとってこれらの情報から誤っている設定項目を特定することは困難である ．そこで，我々はIPネットワーク構築演習支援システムの自動採点機能を開発し た．本機能では，採点結果を視覚的に表示するため，間違った箇所の特定が容易 になる．

Hybrid-P2Pを活用した資料共有システムを開発している。本システムでは各資料 の所在をインデックスサーバで管理しており、検索はインデックスサーバ上で行 う。検索が完了し、目的の資料の所在が分かるとノード間で直接資料が交換され る。検索効率をよく行うためには、ファイル名と併せてキーワードによる検索が 有効である。本研究はノードのディレクトリ構造を利用して、キーワードを自動 的に付与しインデックスサーバに登録する機能を開発する。

大学などの教育機関においてネットワーク技術者の養成を目的とした演習が実施されている．我々はこれまでにNetPowerLab?と呼ぶ，仮想マシンを活用したハンズオンタイプのIPネットワーク構築演習支援システムを開発してきた．このシステムを利用することで，学習者は，実機を用いること無くネットワークの構築演習を実施できる．本研究では，ルーティング情報を学習するための新しいシステムを開発した．この新しいシステムはNetPowerLab?を基盤にしている．本システムでは，ネットワーク上で交換されるルーティングアップデートの内容をリアルタイムに表示する．これにより学習者は，ネットワークを構築しながらネットワーク上の全てのルータのルーティングアップデートを確認できる．

IPネットワーク上で音声、映像、データサービスが混在した統合型通信が増加している。音声や映像などのリアルタイム性のある通信はジッタやパケットロスに敏感であるため、ネットワークを設計する際にはトラフィックの流れ方や量を分析し、それに合ったQoS制御方式を検討する必要がある。 本研究では、設計したネットワークの事前検証を支援するため、仮想マシンを活用した仮想環境上でQoSネットワークをエミュレーションするシステムを設計した。本稿では、仮想環境上でQoSをエミュレーションするための要件を整理し、それを満たすために設計した機能について述べる。

本研究では，SNMPを用いて収集した情報を基に，IPネットワーク構成情報を推定する． 現在，ネットワーク管理等の目的で，tracerouteやOSPF-LSDBを用いたIPネットワーク構成情報 の推定がされている．しかし，どの推定方法も一長一短がある．そこで，本研究では，既存の推定方法の問題点を踏まえ，SNMPで収集したMIB-IIを用いての推定を試みた．本稿では，推定の際に使用するMIBと推定方法について述べる．さらに，評価実験として，実験用の実ネットワークと，推定結果を基に作成した仮想的なネットワークを比較する．そして，その実験結果と今後の課題について言及する．

The purpose of this study is to give students a hands-on practical learning for IP network skills in campus cloud computing. In this study, we have developed a handson practical learning system to learn IP network skills on general performance PC. Unlike existing network simulators, this system does not require students to learn scripting language to make scenarios. This system enables students to practice network skills freely and virtually on just one PC. We have implemented this system by applying the virtual Linux environment User Mode Linux. This system consists of virtual routers, virtual hosts, and operating functions. This system is primarily intended to practice operation of routing on virtual routers. In this research, we developed new system that was designed on client server model for web services. To evaluate, this system was experimented. The results indicated that this system had sufficient performance for introductory network courses for undergraduate students.

近年，コンピュータネットワークの普及により，大学等の教育機関においてネットワーク技術者の養成を目的とした教育が実施されている．我々はこれまでに，一台のサーバ上に複数の仮想マシンを動作させ，それらを相互に接続することでネットワークの構築演習を可能とするIPネットワーク構築演習支援システムを開発してきた．IPネットワーク構築演習支援システムでは，実機を用いることなく，IPネットワークの構築演習が実施できる．さらに，クラウド環境を用いることで，構築演習環境をWebアプリケーションとして提供できる．今回，複数の学習者が一つのネットワークを構築する協調構築演習の要件について整理し，協調構築演習を実施可能なシステムを検討する．

本研究では，これまでに開発したIPネットワーク構築演習支援システムの各機能を基に，学内クラウド環境で運用可能なシステム(以下，本システム)を新たに開発する．本システムはWebアプリケーションとして動作するため，学習者は学内ネットワークからWebブラウザを用いることで，手軽にネットワークの構築演習を実施できる．

本研究では，仮想環境上のネットワークを学習者同士で構築可能なシステムを設計し， そのシステムを用いた協調演習を検討する．本システムは協調演習時の学習行動を考慮したシステムで構成され，学習を支援する機能を実装することで協調演習が可能になる．

本研究では，仮想スイッチを用いたIPネットワークの構築演習を開発した．L2スイッチを利用する演習としては，VLAN，スパニングツリープロトコル（以下，STP），L2セキュリティの設定を行うことを想定している．そこで，本研究では，これらの演習を可能にするL2スイッチを開発した．本システムより，学習者が実施することができるIPネットワークの構築演習が増加する．

本研究では，仮想マシンで構成される仮想的なネットワーク上でQoSを検証するためのシステムについて検討した．仮想ネットワーク上でQoSを検証するための要件を述べ，本システムと既存のQoS検証手法の違いを比較する．

本研究は，仮想マシンによるIPネットワークのテスト環境の提供を目的とする．本環境では，学生が製作したネットワーク関係のソフトウェアのテストなどが可能となる．今回は，テスト環境の設計と試作について述べる．

The purpose of this study is to give students a hands-on practical learning for IP network skills. This system enables students to practice network skills freely and virtually on just one PC. We have implemented this system by applying the virtual Linux environment User Mode Linux. This system consists of virtual routers, virtual hosts, and operating functions. In this research, this system is primarily intended to practice operation of routing on virtual routers. To evaluate, this system was experimented. The results indicated that this system had sufficient performance for introductory network courses for undergraduate students.

オブジェクト指向Web アプリケーションサーバの一種であるZope を用いて、BIX 用画像ファイルサーバを構築した。BIX 画像用サーバの機能とBIX 用ポータルサイトの構築を行った結果について報告した。

高精度GPS によって計測した地形データを可視化し、圃場管理に利用するシステムを構築した。画像にはSVG を用い、階層化した画像によって様々な表示を可能とした。

オブジェクト指向のWebアプリケーションサーバを利用し、農業情報に適した画像データベースの構築に関する提案を行った。静止画像に対して、サムネイルの自動生成機能等を実装したプロトタイプを構築した。（英文）

果樹栽培を支援するためのウエアラブルコンピュータの利用方法について提案した。本システムは、超小型ディスプレイとビデオカメラ、音声入出力装置、マウス、小型キーボード、および超小型PC本体から構成される。

診断用画像に対して階層型符号化を施し、領域と階層の組合せに対して優先権を付与する方式について提案した。ルータは優先権に基づいてパケットを転送する。実験から注目領域を優先的に転送できることが分かった。

果樹育成に関する知識をデータベース化し、利用者のレベルに応じて、必要な情報を取り出して表示するシステムを提案した。今回は、果樹育成に関する知識の体系化について報告した。

診断用画像に対して階層型符号化を施し、領域と階層の組合せに対して優先権を付与する方式の実装について報告した。階層型符号化にはPVHを利用し、領域分割方法の適応を試みた。実験から目的通りの動作を確認した。

果樹栽培を支援するためのセンサー類を相互接続するネットワークの構成とサーバによるデータ交換の方法および状況を判断する機能について提案した。一部について実装し、実験した結果を示した。

農作物栽培を支援するための各種センシング方法とそれらからデータを利用者間で円滑に交換するための仕組みであるBIXの概念について報告した。BIXはXMLによって設計される情報交換に手段である。

農作物栽培を支援するための各種センシング方法の提案と特にそれらの連携に関する戦略について提案した。さらにXMLの活用によって測定データを相互に利用するための仕組みであるBIXの概念について報告した。

本研究課題では、Webアプリケーションに対するSQLおよびNoSQLインジェクション対策の学習を支援する、仮想マシンを活用した実践的な演習システムの開発研究を目的としている。本システムによって、学習者は安全かつ手軽にセキュリティ対策の実践的な演習が実施できる。さらに本システムを用いることで、学習者は一人で対策演習と攻撃演習を実施することが可能となるため、身近に他の学習者がいない環境、たとえば対面による実習・演習が困難な場合や学習者が自宅で自己学習として演習を行う場合でも、対策側と攻撃側の双方の演習が可能となる。 R３年度では、個々のWebページを対象とするSQLインジェクション対策演習機能を実装した。開発実装した機能によって、標準的なPC上で、個々のWebページに対するSQLインジェクションの攻撃手法とそれに対する対策手法の演習の実施が可能となった。 具体的に実装した機能は、 SQLインジェクション対策演習部・攻撃演習部である。攻撃演習部ではWebアプリケーション脆弱性診断ツールであるBurp Suiteを組み込むことで、脆弱性の発見方法の学習を可能とした。脆弱性を発見したら、OWASO ZAPとSQLmap等を使って、実際にSQLインジェクション攻撃の体験が可能となった。対策演習部によって、対策手法の解説ページを参照しながら、Webアプリケーションのソースコードの編集を可能である。 情報系の学生20人を対象とした利用実験を実施した。事前テストを実施後、被験者を座学によって学習するグループと、本システムを使用して学習するグループの2つに分け、学習後に事後テストを実施した。その結果から、開発した機能の有用性を評価した。実験の結果、座学で学習したグループと比較して本機能を利用したグループの方がより有意に点数が高いことが確認できた。

生産物が水面下にある魚類養殖では，生産管理や成育状態を日常的に把握することが困難となっている。そのため，きめ細かい生産管理や効果的な養殖技術開発のためにICT導入による高度化が強く求められている。本研究では魚類養殖の成育・生産を効率化するシステムの構築を目指して，非接触で生簀内養殖魚の収容尾数や，成育状態を計測する基盤技術開発に取り組む。 【個体数計数システムの開発】養殖クロマグロを対象として遊泳する養殖魚魚群をワイドコンバージョンレンズを装着した小型民生品カメラを用いて20m直径の大型生簀内のほぼ全景を捉えることができた。セマンティックセグメンテーション（画像内の全画素にラベルやカテゴリを関連付ける深層学習アルゴリズム）を適用して個体画像を識別したところ，およそ5000枚の教師画像データデータセットを個体画像と背景画像の比率7:3で学習させることにより80%の検出性能を出すことができたが，システムのロバスト性を向上させるためには，個体を移動追跡するアルゴリズムを見直し，実海域での個体同士の重なりや懸濁物等のノイズ等の識別を確実化することが求められことがわかった。 【行動計測による養殖魚状態評価技術開発】個体計数システムの開発では，連続する画像間での同一個体の移動ベクトルをKC法を適用することにより推定することが可能となる。このベクトルの状態を分析することにより，養殖環境や適切な給餌量を評価することが実現できる可能性がある。そこで，連続する移動ベクトルのノルム差D，交差角θ，渦度に類似した指標ωの評価指標を算出し，養成環境の違いがこれらの指標に差をもたらすか分析した。給餌時と非給餌時では，θの値で有意な差が確認され，台風接近による風浪の激しい環境では，ノルム差やωの値に有意な差が現れた。養殖魚の飽食度合いや，波浪によるストレス度合をこれらの指標により評価できることが示唆された。

In this research, we developed a system for supporting learning of network security through offensive and defensive battle exercise. In this system, two learners are divided into an attacker and a defender. This system utilizes free software and virtual machine technologies so that low-cost and safe exercise is possible. In addition, since the user interface is a browser, learners can easily conduct exercise. Through experimental evaluations, we confirmed that our system can support learning of network security.

In this research, I developed functions to realize cooperative practice of IP networks by learners and software agents in practice of IP networks utilizing the virtual Linux environment. Even without cooperative learner, learners are able to practice of IP networks as cooperative practice. The main functions that I have developed are 1) software agent function with IP network construction skills, 2) ability to make collaborative practice with software agent, 3) automatic scoring function. I also developed an automatic problem creation function. From experimental results, I confirmed that each function of this system is operating normally, and it is possible to collaborate with software agents.

In this study, I developed function that automatically scores IP network settings for IP network practice system. And I developed function that supports to make problems for teachers. I introduced this system to classes of undergraduate students. As results, I found that this system had usability for IP network practice classes. And I timed automatic scoring. This system allows students to engage self-guided IP network training and testing.

I developed a sensor data exchange system with hybrid type P2P in order to share sensor data. This system consists of an index server and some nodes. The index server manages meta data of sensors. The node manages data of sensors. First, I implemented the index server and the node. Then, I implemented a sensor managing function and a mini node function. I found in that experiment this system can share sensor data easily.

本研究課題では、果樹園を対象としてセンサネットワークシステムおよびセンシングデータ閲覧システム、潅水コントロールシステムを開発した。センサネットワークシステムはバッテリ動作を目指し、省電力化を図るため有線と無線によるハイブリッド構成を用いた。また、センシングデータ閲覧システムはSVG を用いて圃場マップを描画することで、GIS 機能を持たせた。また、Google カレンダーを用いた潅水制御インタフェースも開発した。

(1)NASA/AERONET放射計等による大気エアロゾル地上観測、(2)SPMのサンプリング計測と化学分析、(3)エアロゾル放射特性とSPM値の関係解析、(4)地上と衛星データの併合によるエアロゾルマップの導出等に関し、継続的に研究実績を挙げている. (5)2008年5月にNIESライダーネットワーク機器を設置し、(6)エアロゾルの発生から拡散までの解明、(7)総合的なデータ解析手法の実装、(8)モデルシミュレーションと実データの融合を実現した.