電気学会論文誌C(電子・情報・システム部門誌)
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
118 巻, 11 号
選択された号の論文の25件中1~25を表示しています
  • 宇宙からの地球環境観測最前線-リモートセンシング-
    安岡 善文
    1998 年 118 巻 11 号 p. 1543-1546
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 人工衛星による地球環境観測の現状と将来
    下田 陽久
    1998 年 118 巻 11 号 p. 1547-1550
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 地球環境観測とモデリング
    住 明正
    1998 年 118 巻 11 号 p. 1551-1554
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 景山 陽一, 西田 眞, 内海 富博
    1998 年 118 巻 11 号 p. 1555-1561
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    This paper proposes a new method for the automatic extraction of the lineaments on the SAR (Synthetic Aperture Radar) image data. It is well known that the lineament patterns are the important feature reflect the structures of the subsurface. A lineament map has been drawn by the fieldworks of experts and the analysis of the enhanced image data using visual interpretation. It is therefore indicated that automatic extraction of the lineaments on the SAR image data is useful and effective for understanding the geological structure in the study area.
    This paper generates the edge data which contains the lineaments using some operators and template matching at first. Next, the emphasis of edge with the scored method is carried out. It is assumed that the edge pixel having high scored value can be much related to the lineaments. Finally, the trace and connection of the edge is conducted with the above assumption. As a result, it is confirmed that the results with the proposed method correspond to the lineament map drawn by experts. Therefore, it can be concluded that the proposed method is the effective technique for automatic extraction of the lineaments on the SAR image data.
  • 奥谷 巖, 呉 豪翔
    1998 年 118 巻 11 号 p. 1562-1569
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    In this study developed is a new method for land-cover classification of remotely sensed data. The proposed method bases its modeling structure on so-called category decomposition principle, estimating categorical proportions of a mixel rather than assigning only one category to it. Classification algorithm is established in the framework of fuzzy linear regression analysis so that observed radiance data of the subject mixel may be contained, at least to a predetermined degree, in the estimated fuzzy set of radiance which is accomplished by the weighted sum of category specific spectra expressed by fuzzy numbers. Validity tests using Landsat TM data show that the proposed method substantially outperforms three existing methods compared, the quadratic programming method based on category decomposition idea, the linear discriminant analysis method and the maximum likelihood method, by up to 40% in terms of root mean square error of estimation and also the proposed method has a kind of robustness in that it can make relatively accurate estimates compared to the existing methods even though inexact categorical characteristics of spectra are input to the model.
  • 景山 陽一, 西田 眞, 肥田 登
    1998 年 118 巻 11 号 p. 1570-1576
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    This paper presents the results of using local fractal dimension to estimate the flood plain in the Amazon on JERS-1 SAR (Synthetic Aperture Radar) image data. The change of water level reaches about 10 meters in the middle and lower Amazon, and then large area may come under water at high levels. The area, called várzea, has a lot of relation to human activities. Study area was selected at the west of Parintins-city, Brazil and a case study was conducted.
    To estimate the várzea area, a new analysis method on JERS-1 SAR image data is proposed. The method consists of three steps. First step is the estimation of three classes. They are the water areas, forest areas and intermediate class between two classes. Second step is the estimation of mixel (mixed pixel) by means of edge information. Third step is the extraction of the várzea area and other classes from intermediate class using the local fractal dimension. Estimation of the class is carried out with the fuzzy estimation method. As a result, it is confirmed that the estimation result for the proposed method matches the knowledge obtained in the ground survey. Therefore, it can be concluded that the proposed method is the effective and practical technique to estimate the várzea area in the Amazon on JERS-1 SAR image.
  • 森 俊介
    1998 年 118 巻 11 号 p. 1577-1583
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    An assessment of postponing the decision making on the global warming mitigation options is described. Recently global warming issues are well concerned by the academic and the policy making fields. Even though everyone agrees with the reduction of greenhouse gases (GHG), there are still disputes on how and when to implement. There still remain so many uncertainties on the future global warming, especially on its future societal and economic impact, that some emphasize the importance of early actions while the other insist the economic efficiency of postponing the decision. Some integrated assessment models have dealt with the future uncertainties to explore the most preferable policy options, but few assessed the impact on the technology strategies. In this paper, the author extends the integrated assessment model MARIA (Multiregional Approach for Resource and Industry Allocation) to deal with the decision making under future uncertainties employing the scenario optimization technique. Changing the branching point of decision making, the effects of postponing are evaluated. The simulation results show us that postponing the decision making requires us to implement the carbon free technologies, e.g. nuclear power and biomass utilization, earlier when the atmospheric carbon concentration is stabilized at 550ppmv.
  • 田中 久弥, 長嶋 祐二, 井出 英人
    1998 年 118 巻 11 号 p. 1584-1589
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    Up to now, Communication aid systems which use biosignals for physically handicapped have been proposing. On the other hand, In the field of virtual reality, man-machine interface which use biosignals is being watched with keenest interest. This paper described the analysis and discrimination method of intentions using MRCP.
    The evoked potentials called MRCP(movement-related cortical potentials) are evoked when a man move with intention. Especially, RP (readiness potentials) that locate just before movement are not evoked without intention. So it thinkable that RPs are reflected peculiarly in the readiness. We approach RPs. We study that measure MRCPs in playing JANKEN and analyze frequency in the RPs. We study that do principal analysis in GU, CHOKI and PA. We built JANKEN discrimination model.
    As a result, We found principal factors in 4Hz and 8Hz frequency spectrum. The discrimination system of JANKEN which built from the principal factors discriminated GU 62.5%, CHOKI 50.0% and PA 75.0% in unknown RP data. We analyzed the relations between movement forms and RP. And suggested potentials of building the 3mode intention communication system.
  • 景山 陽一, 西田 眞, 豊福 克也
    1998 年 118 巻 11 号 p. 1590-1595
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    This paper presents a new fuzzy estimation method for the class mixture proportion considering the number of component class in the mixed pixel (mixel) on a remote sensing image. The method consists of two steps. First step is estimation of the number of the landcover classes in the mixel. Second step is estimation of the class mixture proportion based on the above class number. The distribution of the training data was defined as a fuzzy set on a spectral space. It was assumed that the spectral characteristics of the mixel was regarded as a linear function of the reflection level of the pure pixels corresponding to the component class. Both the number of component class and the class mixture proportion were estimated by the fuzzy simplification reasoning method in the proposed method.
    The simulation results indicated the reasonable distinction as the number of component class. That is, the threshold value for two, three and four classes were obtained as 0.90, 0.97 and 0.97, respectively. It was observed that the proposed method gave low TRMS (the Total Root Mean Square error) to the simulation data that was produced by the random sampling data of each training class. Therefore, it was confirmed that the proposed method was an effective and practical technique to estimate the class mixture proportion in the mixel.
  • 竹村 淳, 伊東 正安
    1998 年 118 巻 11 号 p. 1596-1602
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    Today, information obtained a medical image is important for a diagnosis. Especially, ultrasonic image has the merits of the simplicity of the machine, low cost, and safety. On the other hand, ultrasonic image is sometimes poor in quality because of much noise. Therefore, the diagnosis by ultrasonic image has been depended on the judgement of skilled doctors. It is necessary to develope a method of an automatic diagnosis that is based on a qualitative evaluation.
    In order to extract automatically a region of an internal organs or a region of focus from medical ultrasonic image, it is neccesary to detect a defferrence of some quantitative features between a region as a purpose and other regions. Usual methods for regional extraction of ultrasonic image was based on features defined from gray-level or statistical characters of speckle pattern. In those methods, various features were defined as applying texture analysis. But effective feature, that is used to extract a tissue region as a purpose, is seleceted empirically.
    In this research, we propose a mothod to select an optimal set of discriminating functions, that is composed of effective features to extract tissue region by use of genetic algorithm (GA).
  • 伊藤 欣二郎, 寺崎 健, 松井 照明, 萩原 賢一
    1998 年 118 巻 11 号 p. 1603-1610
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    The AI system is expected to support the user's decision work and be able to update knowledge bases to meet changes caused by change in equipment easily. AI engineers, called knowledge engineers, often have to acquire empirical knowledge through interviews with skilled operators and other means. The technical difficulty is known as the bottleneck in knowledge acquisition. Thus, the characteristic and the limit of the AI system are recognized through the actual construction. AI system has shifted from a AI independent system which aims at human expert's substitution to the subsystem to improve user's benefit and convenience.
    This paper reports about the tool for knowledge acquisition and its application to the insulation diagnosis of the high voltage rotating machines.
  • 高田 等
    1998 年 118 巻 11 号 p. 1611-1616
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    In this paper an automatic choosing control of sigmoid type is considered for nonlinear systems. Nonlinear terms of the system are linearlized piecewise so as to be able to design the linear optimal controllers. These controllers are smoothly united into a single nonlinear feedback controller by automatic choosing functions of sigmoid type. Parameters of these functions are suboptimally selected by minimizing the Hamiltonian with the aid of a genetic algorithm. The resulting controller is a compact form which enables easy implementation and evaluates a stable region in the sense of Lyapunov by a Lyapunov function candidate. Simulation results show that the new controller can improve performance remarkably well.
  • 小原 和博
    1998 年 118 巻 11 号 p. 1617-1622
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    We proposed ways to improve pattern recognition ability by combining several small back-propagation neural networks (BPNNs) [1]. We found that modifying the desired outputs according to the similarity of the input patterns (i.e., increasing desired outputs to similar classes) increases the BPNN outputs for similar classes, thus improving the generalization ability of the modular-net architecture. We evaluated the learning technique using two subfeatures extracted from handwritten digits [1]. This paper proposes a performance-verification method and presents experimental results applying learning techniques to the proposed verification-problems: 4-class, 10-class, and 20-class classification problems using two-dimensional Gaussian distribution data.
  • 永野 宏治
    1998 年 118 巻 11 号 p. 1623-1632
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    This paper investigates a low-velocity-layer (LVL) model for an artificial subsurface fracture examining dispersions of crack-waves. Subsurface fracture is important to extract geothermal energy. The crack-waves are seismic modes propagating along the fracture. The LVL represents a zone in which many small fractures are distributed around a main single fracture. The crack-waves were measured at an artificial subsurface fracture at a depth of about 370m in Tohoku University Higashi-Hachimantai Hot Dry Rock model field, Japan. The Wavelet Transform is applied to describe the dispersion of the crack-waves. I obtain the crack-wave dispersions which are affected by the pressurization of the subsurface fracture. The numerical dispersion curves are also calculated in the LVL model using a matrix method. When the aperture of the subsurface fracture increases by the pressurization, the LVL model shows a good agreement with the crack-wave dispersion. The LVL model is, however, insufficient for the crack-wave dispersion detected at the beginning of the pressurization.
  • 奥野 秀樹, 加瀬 渡
    1998 年 118 巻 11 号 p. 1633-1638
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    Recently, a design of model matching and adaptive control systems under the presence of measurement noise is reported by employing a nonminimal-order state observer. But this method can not be used for the case where the plant has the deterninistic disturbances. In this paper, we will show that the idea is also applicable to the case where the plant has both the measurement noise and the deterministic disturbances. For this purpose, we will employ a direct method of adaptive control where the controller parameters are estimated.
  • 川端 邦明, 石川 達也, 淺間 一, 嘉悦 早人, 遠藤 勲
    1998 年 118 巻 11 号 p. 1639-1645
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    In this paper, we propose an autonomous teleoperation system with IDCs (Intelligent Data Carriers). An IDC is a device to enable the working environment to work as an agent of the system. In conventional teleoperation techniques, the load of the task execution is shared by the operator and the robot as a teleoperated target system. However, the proposed teleoperation system allows the load to be shared by the operator, the robot and the IDC as a local information strage of the environment. By using IDCs for the teleoperation system, the required sensing ability of the robot can be reduced, because the robot can take the information about local working environment from IDCs which are placed in the remote environment. This implies the effectiveness of implicit collaboration among the operator, the robot and the remote environment. In this paper, we make some experiments with the real system through the Internet to show the effeciency of this system.
  • 柴田 昌明, 恩田 寿和, 田村 公良
    1998 年 118 巻 11 号 p. 1646-1655
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    The paper describes a novel approach of feature extraction for the fast object recognition based on Model Based Matching method (MBM method) in computer vision researches. The proposed approach is aimed to achieve the recognition in video-rated processing. In general the MBM method is known and utilized as the robust method against the variation of the luminance of target objects and the brightness of their circumstance. On the conventional process in the MBM method, however, it spends more than hundreds [msec] from taking the image of the objects to detecting their positions and orientations because of the large amount of calculation. In our approach, by introducing the Model made of central positions of circles only, quite novel process are structured, which is composed of just two steps to extract the positions. The first is to obtain the differential image by applying the filtering mask feasibly designed in our method. The second is to extract the central positions of circles from the differential image. The validity of our method is confirmed in the experimental results.
  • 吉廣 晃, 小島 隆夫, 園田 敏勝, 辻 輝生
    1998 年 118 巻 11 号 p. 1656-1662
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    Recently it seems that there are almost no multipliers using Hall elements. This fact mainly comes from the magnetic nonlinearity such as magnetic saturation and hysteresis of a magnetic core used in order to increase the gain of the magnetic flux path.
    This paper shows realization of a high precision multiplier using a Hall element with magnetic flux control. Moreover, the relation between the precision of the flux control and the precision of the multiplier is clarified. As a result, it is shown that the precision of the proposed multiplier is considerably higher than that of a existing IC multiplier.
  • 東 秀樹, 原田 成光, 園田 敏勝
    1998 年 118 巻 11 号 p. 1663-1669
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
    For measurement of nano-voltages it is required to evaluate the noise which comes from a Zener diode and a lithium battery which will be used here.
    This paper discusses the problems which arise when a commercial digital multimeter (D.M.M.) which has nominally resolution of 10nV is used for some measurement. That is, the magnetic fields in a usual test room which is 10-9 to 10-7 T with 60Hz and also the magnetic field which is 10-7 T and comes from the power supply part of the D.M.M. through the transformer will disturb nano-volt measurement. These magnetic fields possibly induce a 10nV in the measurement circuit of the D.M.M. This fact recommends the magnetic shielding of the measurement circuit. Moreover the D.M.M. shows 10∼20μV when the input terminals are short circuited. This fact is very a serious problem for measurement of nano-volt. We observed that linearity is not hold lower than 800nV using a single frequency sinusoidal voltage input.
    In order to resolve the problems mentioned above, a measured voltage is pre-amplified by a low-noise amplifier with magnetic shielding before it is input to the D.M.M. As a result, the lowest linearity improved from 800nV to 10nV is obtained.
  • Heywood Absaloms, Takehiko Tomikawa
    1998 年 118 巻 11 号 p. 1670-1671
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 童 衛青, 辰巳 昭治, 志水 英二
    1998 年 118 巻 11 号 p. 1672-1673
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 茂中 義典, 藤田 得光, 渡辺 隆男, 安田 恵一郎
    1998 年 118 巻 11 号 p. 1674-1675
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 大西 義浩, 山本 透, 沖 俊任, 兼田 雅弘
    1998 年 118 巻 11 号 p. 1676-1677
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • Ikuo Yamashita, Shigeyuki Seikai
    1998 年 118 巻 11 号 p. 1678-1679
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
  • 島田 賢, 田中 久弥, 井出 英人
    1998 年 118 巻 11 号 p. 1680-1681
    発行日: 1998/11/01
    公開日: 2009/10/02
    ジャーナル フリー
feedback
Top