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
117 巻, 10 号
選択された号の論文の29件中1~29を表示しています
  • 坂上 勝彦, 輿水 大和
    1997 年 117 巻 10 号 p. 1339-1344
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
  • 笹間 宏, 鵜飼 正人, 太田 勝, 宮本 岳史
    1997 年 117 巻 10 号 p. 1345-1354
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Continuously Scanned Image (CSI) gives convenient image data for inspection of elongated facilities in railway, such as tracks, rolling stocks and so on. We have developed a system named ConSIS to take the CSI at a high data rate of 40M pixel/sec, which is composed of a line-sensor camera and a data-recorder. As the result of field tests taking the CSI of tracks, rolling stocks and tunnel walls, feasibility of an inspection system using the CSI was confirmed.
    Additionally, we have developed a prototype image database system to examine effective use of the CSI, in which some useful functions such as normalizing a scale of the CSI, scrolling the continuous image, zooming up a fine image for observation, are implemented. Further, we have developed a method to extract cracks on a concrete surface for automatic inspection using the CSI and a computer.
  • 塚田 敏彦, 中野 倫明, 山本 新, 松尾 啓志, 岩田 彰
    1997 年 117 巻 10 号 p. 1355-1361
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    A method for measuring the shapes of 3-D objects, of which surfaces are composed of various reflection value regions, has been developed. The 3-D coordinates of the object sections are obtained by triangulation with the light-section method using a slit light source and a TV camera. A TV camera having an expanded dynamic range is employed to reduce the influence of uneven reflection surfaces. The TV camera expands a dynamic range by combining plural images captured at different shutter speeds.
    A trial 3-D shape measurement system has been produced using the developed method, of which ability and effectiveness of this method were comfirmed by experiments.
  • 佐藤 和也, 山澤 一誠, 八木 康史, 谷内田 正彦
    1997 年 117 巻 10 号 p. 1362-1369
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Described here is a guide mobile robot system for following a user and navigating the user to the desired position in the environment such as museum or exhibition. The system acquires an omnidirectional information around the robot, in real-time, with multiple sensors such as the omnidirectional image sensor HyperOmniVision, omnidirectional ultrasonic sensors, etc. Since HyperOmni Vision observes a 360 degree view around the robot, it can observe a global and precise azimuth information of features, however, it is difficult to estimate the precise location of a moving object (user) from azimuth information. On the other hand, The system has active sensors such as omnidirectional ultrasonic ranging sensors and obtains range data around the robot. However, the azimuth resolution is poor and it is not suitable for the object recognition. Therefore we integrate both sensors'merits. HyperOmni Vision identifies and tracking the user by using the azimuth and color information. Furthermore, the location of the robot is estimated by HyperOmni Vision. The robot follows and navigates the user with the constant distance between the user and the robot by using azimuth information (HyperOmni Vision) and range information (Ultrasonic Sensors).
  • 土橋 外志正, 長田 典子, 眞鍋 佳嗣, 井口 征士
    1997 年 117 巻 10 号 p. 1370-1376
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Visual simulation technique using CG has attracted wide attention in many fields. In this paper, we propose a method of modeling and visualizing pearls to implement a pearl visual simulator. Pearls manifest a very specific optical phenomenon. To investigate this feature, we propose a physical model considering the multiple reflection in spherical bodies. In our previous work, interference component, mirroring component and texture are considered as the principal factors for realistic representation of pearl. This paper proposes two additional factors to improve the visualization quality, which corresponds to internal blurring of an incident light and partial coherent interference. We developed a fast algorithm using subsurface reflection distribution to represent a blur. Experimental result shows the physics-based-modeling of internal blurring in subsurface multi-layer structure of a pearl and a partical coherent interference model are effective for high quality visualization.
  • 鈴木 香織, 斎藤 英雄, 小沢 慎治
    1997 年 117 巻 10 号 p. 1377-1383
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Sensing of human face orientation is vitally important tasks for applications to face to face teleconferences and so on. Most of methods for estimating face orientation from image processing demand extracting face features. However, feature extraction is much affected by environment inclusive of lighting and background, and some of them need manual operations. This paper proposes a robust method for estimating face orientation from shading images without feature extraction. First we obtain the 3-D shape and a shading image of object face by the equipment incorporated with a FG range sensor. Then head images with various postures are synthesized in advance. Next we compare an object image with synthesized images. The head-posture of the synthesized image having highest correlation with the object image is determined as the estimated posture. As a result of experiments using image sequences as the input, synthesized image sequences changed the face orientations following the change of face orientaions of input image sequences.
  • Seong Ik CHO, Hideo SAITO, Shinji OZAWA
    1997 年 117 巻 10 号 p. 1384-1390
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    A shape recovery problem of book surface using two shade images under the fully perspective environments is discussed. In order to simplify the problem, the whole recovery process is divided into three sequential and explicit steps: preprocessing, apparent shape recovery, and generation of ortho-image. The separation of albedo and shading with reduced effect of interreflections is done in preprocessing step by applying image processing and a phenomena-based model. Implicit equations governing the shading and observation have been transformed into explicit ones having minimized number of unknown parameter. A direct and unique recovery become possible by combining the transformed ones and the recurrence relation. A feed-back recovery process is implemented as a practical algorithm which overcomes self-shadows. The results of simulations and real experiments show the properness and acceptability of the proposed approach and the implemented algorithms.
  • 松本 敦行, 斎藤 英雄, 小沢 慎治
    1997 年 117 巻 10 号 p. 1391-1397
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper describes a new method for reconstructing the structure of a skin surface replica from three shading images taken from three different lightings. Since the shading images include specular and inter reflections, the conventional photometric stereo method is not suitable for reconstructing its surface structure. In the proposed method, the evaluation function of the surface shape is defined, then the structure is reconstructed by optimizing the evaluation using simulated annealing. The experimental results from both synthetic images and real images demonstrate that proposed method is more effective for shape reconstruction from shading images which include specular and inter reflections.
  • 橋本 栄一郎, 木原 均, 宮本 幸三, 東原 稔, 人見 正明
    1997 年 117 巻 10 号 p. 1398-1403
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    With advances in semiconductor technologies, applications for image recognition technologies have found their way from industry to business and consumer fields. We have developed a Japanese chess recording system that automatically records and stores real-time chess moves using a CCD camera located above the chessboard. The system detects chess moves and specifies chess pieces using image recognition techniques and chess rules, and graphically displays the chessboard status on a personal computer. In the past, ideal lighting was necessary to implement the image recognition techniques. In order to use this system under nonideal conditions, for example a general life environment, we have developed the algorithm that is robust to unstable illuminance.
  • 高松 亮, 高田 康弘, 佐藤 誠
    1997 年 117 巻 10 号 p. 1404-1411
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    In this paper, we propose a method to estimate posture of human head, which is suitable for constructing vision-based human interface. By the proposed method, the position and the direction of the head are estimated from the image of singular marker put on the head of a user. As the feature, moments of image up to second degree are used. Posture is estimated by estimation system which consists of observation model based on local analysis and compensation function. Our method has the following characteristics important for posture estimation method as an element of general use human interfaces. Firstly, it does not constrain user, since it is vision based method. Secondly, a user only have to put singular marker on the top of the head, and thus easily puts on and takes off the marker. Thirdly, because the features to represent the marker are moments up to second degree, we can expect considerable robustness to noise and less computational cost. The experimental results shows validity of our method.
  • 王 耀東, 舟久保 登
    1997 年 117 巻 10 号 p. 1412-1417
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper studies an application of geometric primitives extracting methods by genetic algorithm (GA) to detection of road sign boards from scene image. We attempted to simultaneously find out the primitives using newly considered equivalence genes. To extract certain type of geometric primitives by GA has been already studied. However, there is hardly any research which applies the multimodal function optimization problems to simultaneous detection of geometric primitives by GA. In this paper, we propose a new approach with the functions such as observation of convergence, promotion of convergence, confirmation of convergence and maintenance of multiple subpopulations by equivalence genes to solve the above problem. We also confirmed its practical usefulness through experiment.
  • 綾木 之裕, 上町 新也, 石橋 武, 白井 良明
    1997 年 117 巻 10 号 p. 1418-1424
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper describes a trinocular vision system to detect heavy machineries such as mobile cranes approaching power transmission lines for avoiding contact accidents. Because the monitoring area is large, a passive measuring method is desirable. We have developped trinocular vision to acquire range data more reliably than conventional binocular vision. In order to find the correspondence of features in three images, we use a disparity histogram for matching candidates in a local area. By this way, the disparities of horizontal or vertical edges can be determined as well as reliable correspondence is obtained. We apply this trinocular vision to the monitoring system which is attached to a tower for power transmission lines. The system monitors a prespecified space to detect objects approaching power transmission lines. Experiments were carried out for real situations and proved the system is satisfactory.
  • 野引 敦, 成瀬 央, 立田 光廣, 藪田 哲郎
    1997 年 117 巻 10 号 p. 1425-1433
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper proposes a novel outdoor facility measurement system. The purpose of this system is to enable survey work on roads to be undertaken both safely and efficiently. This system employs a stereo measurement method and measures facility positions by integrating stereo measurement data from multi-viewpoints whose relationship is unknown. The system involves two processes which consist of recording images in the field and then analyzing those images indoors. The system is set up in the field and stereo images are recorded in a digital camera from several viewpoints around an intersection. The image analysis is divided into three parts. First, the facility positions from each viewpoint are obtained by a stereo measurement method with the aid of an operator. Second, the correspondences between the stereo measurement data from the multi-viewpoints are obtained. Finally, the facility positions and the relationship between the viewpoints are estimated. The second and final stages of this image analysis are performed automatically. This paper presents highly accurate measurement algorithms for this system, which reduces the amount of assistance required from an operator, and describes facility measurement for some actual intersections.
  • 平山 博史, 福山 裕三, 西村 敏博
    1997 年 117 巻 10 号 p. 1434-1442
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Impedance matching condition was investigated by an equivalent electrical circuit model which simulated circulation at the region of femoral and tibial arteries. There is a terminal resistance that is known to minimize the blood flow impedance. The calculated results exhibited that the minimum impedance was influenced by the elastic modulus and the radius of artery. The present results afford important information for a plantation of vascular graft as a replacement of sclerotic artery.
  • Hirohumi Hirayama, Toshihiro Nishimura, Yuzo Fukuyama
    1997 年 117 巻 10 号 p. 1443-1449
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Theoretical analysis was performed on the carrier-mediated solute transport across the biological membrane based on optimal control principle that minimizes the rate of changes in concentrations of the species in combination with energy to drive the system. Experimental data for temporal change in carrier mediated transport could be simulated by the present model with regulating weighting coefficients. Changes in concentrations of the species due to the increase in rate constants could be explained by the compensatory action of the optimal control mechanism. The present study must be available for evaluation of the optimality in carrier mediated membrane transport.
  • 成田 憲一, 竹内 伸直, 中鉢 憲賢, 本間 規泰
    1997 年 117 巻 10 号 p. 1450-1455
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    The variational signals of atmospheric electric field under fair weather has been observed both at Aobayama in Sendai and Tsukidate. This signals are measured using slow antenna type electric field meter which is composed of a metal plate electrode and a CR integrator and recorded using a portable pen-recorder. The field meter of this type is suitable to measure relatively rapid variational field signals. We have observed the characteristic variational signals with a few minutes period in a breeze under fair weather. About these signals we examined the comparison of naked metal electrode and insulated one, the dependences of electrode's area and of electrode's height. From signals observed at two different positions the time delay of two waveforms are recognized. This means that scale of electric field perturbation is estimated about several hundred meters. The already found atmospheric circulation with a few minutes suggests that the observed variational electric field signals are caused by the movement of electric charge together with atmospheric circulation.
  • 呂 建明, 谷萩 隆嗣
    1997 年 117 巻 10 号 p. 1456-1463
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper presents a design method for robust model matching control of nonminimum phase discrete-time systems. This scheme can robustly control the nominal model in the presence of unmodeled dynamics and achieve the desired model matching simultaneously. Furthermore, the sufficient condition for stabilizing the nominal model in the presence of the unmodeled dynamics is derived and the existence of bounds for all signals is proved. Finally, computer simulation results are presented to illustrate the effectiveness of the proposed method.
  • 向谷 博明, 水上 孝一
    1997 年 117 巻 10 号 p. 1464-1471
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper considers new recursive algorithm to find the positive semidefinite stabilizing solution of H type Riccati equation with small parameter where the solution P depend on the ε. In order to obtain the stabilizing solution of the H type Riccati equation, we must solve the generalized algebraic Riccati equation. Using the recursive algorithm, we show that the solution of the generalized algebraic Riccati equation converges to a positive semidefinite stabilizing solution with the rate of convergence of Ok).
    We also show that for singularly perturbed systems, if the H norm of the transfer matrix function is less than the H norms for the fast system and for the reduced slow system, then the H type Riccati equation has a positive semidefinite stabilizing solution.
  • 天野 亮, 堀口 和己, 西村 卓也, 長田 朗
    1997 年 117 巻 10 号 p. 1472-1478
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    We propose a numerical method of the frequency-weighted model reduction. A model to be reduced (an original model) is a stable SISO discrete-time model which is described by high-order state space equations. We design the reduced-order model so that it can interpolate 1st-and 2nd-order information of the original model at complex frequency points (interpolation points) in the unit circle. The characteristics of the reduced-order model greatly depend on the choice of the interpolation points. The proposed model reduction method is a numerical one which chooses the interpolation points by searching in the unit circle to find the reduced-order model such that L-norm of the reduction error is less than a prescribed value. This method has the following features, which shows that it is an effective numerical method of the frequency-weighted model reduction. i) The reduced-order model is guaranteed to be stable. ii) The procedure of finding the reduced-order model is simple and requires relatively a small amount of computation. iii) The order of the reduced-order model can be controlled by choosing the number ssssssof the interpolation points.
  • 土屋 和広, 武藤 佳恭, 黒谷 憲一
    1997 年 117 巻 10 号 p. 1479-1484
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    A near-optimum parallel algorithm for solving the one-dimensional gate assignment problem is presented in this paper where the problem is NP-hard and one of the most fundamental layout problems in VLSI design. The proposed system is composed of n x n processing elements based on the artificial two-dimensional maximum neural network for (n+2)-gate assignment problems. Our algorithm has discovered the improved solutions in the benchmark problems over the best existing algorithms. The proposed approach is applicable to other VLSI layout problems such as the PLA (Programable Logic Array) folding problem.
  • 唐 厚君, 大久保 重範
    1997 年 117 巻 10 号 p. 1485-1489
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    In this paper, a design of model following control system for linear descriptor system with disturbances is proposed, which extends the state space approach to descriptor system. In the process of extension some constraints such as regular, impulse free, R-controllable and R-observable are set up and investigated. During the design of control input, the fact that the characteristic polynomial of descriptor system is generally no monic is taken into account. Both the physical structure of the system and the physical properties of system variables can be preserved because there is no necessary to make transformation of this system. Based on the theory on poles and zeros of descriptor system, a lemma to satisfy the proposed constraints in this paper is deduced, and the boundedness of inner states is proved with this lemma. The effectiveness of this method has verified by an example of an electrical circuit in which the description is a mixture of dynamic and static equation.
  • 福田 修, 辻 敏夫, 金子 真
    1997 年 117 巻 10 号 p. 1490-1497
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper proposes a pattern classification method of time series EMG signals using neural networks in order to estimate a motion intended by a human operator. To achieve successful classification for non-stationary EMG signals, a new network structure that combines a common back-propagation neural network with recurrent neural filters is used. This network is suitable to express time-varying characteristics of time-series EMG signals. Also dynamics of a terminal attractor is incorporated in the learning rule in order to regulate convergence time. The convergence time is always less than the upper limit of a specified time, so that mental stress of the operator for waitting the convergence of learning can be reduced. In the experiments, the EMG signals measured from four subjects during a series of six motions are used. It is shown from the results that the proposed network can achieve a relatively high classification performance and the learning converges within a specified time.
  • 吉村 宏紀, 清水 忠昭, 井須 尚紀, 菅田 一博
    1997 年 117 巻 10 号 p. 1498-1505
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    The multilayer perceptron called Sandglass type Neural Network (SNN) has the same number of units in input layer and output layer and has less units in hidden layer than units both in input layer and output layer. In this paper we clarified the properties of a noise reduction filter using the SNN. The properties were derived basically by use of the result that the output signal of the SNN could be given by Karhunen-Loeve expansion of an input data matrix. Here, we evaluated the improvement value of signal to noise ratio for the optimum number of hidden units. The noise reduction filter was assured to be effective and stable by the computer experiments using sinusoidal signal corrupted by white noise.
  • 八野 知博, 高田 等
    1997 年 117 巻 10 号 p. 1506-1512
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper deals with an identification method based on the automatic choosing function (ACF) for nonlinear systems. A full data region or a whole domain is divided into some subdomains and the unknown nonlinear function to be estimated is approximately described by a linear equation on each subdomain. These linear equations are united into a single one by the ACF smoothly, and thus the resulting model becomes linear in the parameters. Hence these parameters are easily evaluated by the linear least squares method. Besides the subdomains and the ACF are properly determined by the genetic algorithm which has a high potential for global optimization. Numerical experiments are carried out to demonstrate the effectiveness of the proposed method.
  • 卯木 輝彦, 末竹 規哲
    1997 年 117 巻 10 号 p. 1513-1520
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    In this paper, we propose an autonomous vehicle scheduling schema in large physical distribution terminals publicly used as the next generation wide area physical distribution bases. This schema uses Learning Automaton for vehicles scheduling based on Contract Net Protocol, in order to obtain useful emergent behaviors of agents in the system based on the local decision-making of each agent. The state of the automaton is updated at each instant on the basis of new information that includes the arrival estimation time of vehicles. Each agent estimates the arriaval time of vehicles by using Bayesian learning process. Using traffic simulation, we evaluate the schema in various simulated environments. The result shows the advantage of the schema over when each agent provides the same criteria from the top down, and each agent voluntarily generates criteria via interactions with the environment, playing an individual role in the system.
  • 山下 育男, 下浦 一宏, 青海 恵之
    1997 年 117 巻 10 号 p. 1521-1528
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    Application of semiconductor optical amplifiers to wavelength division multiplexing systems in a 1.3-μm wavelength band have been studied. Measured characteristics of gain and noise figure of the SOA having small dependency on polarization states confirm that it is applicable to the repeater amplifier and the pre-amplifier. When it is used as the pre-amplifier of a non-repeated two-channel 10-Gb/s system an allowable line loss of 32dB is attained with the aid of the other SOA used for the post-amplifier. Measured bit error rate of a system with one repeater amplifier is well simulated from the individual characteristics of the SOA and the optical receiver. Using the similar technique, it is predicted that more than ten amplifiers can be applied in the system with cascaded repeater amplifiers on the assumption that the polarization dependency of the gain is ignored.
  • Natsuo Tanaka, Masayuki Fujita
    1997 年 117 巻 10 号 p. 1529-1534
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
    This paper is concerned with a new type of robust tracking control scheme for robot manipulators using SP-D (Saturated Proportional and Differential) feedback loops. In this paper, the control scheme based on quasi-natural potential is generalized from set-point regulation to trajectory tracking by exploiting an effective auxiliary variable which consists of a saturated positional error and a velocity error. Without any uncertainties in the manipulator model, the generalized tracking SP-D control scheme ensures globally asymptotic stability. In the presence of parametric model uncertainties, the tracking SP-D control scheme is improved in order to be robust against the uncertainties. The proposed robust tracking SP-D controller is able to guarantee globally uniformly ultimate boundedness theoretically. Experimental results on a two-link direct drive manipulator with the robust tracking controller are in agreement with the theoretical analysis and show the effectiveness of the control scheme.
  • 竹内 伸直, 中鉢 憲賢, 成田 憲一, 本間 規泰
    1997 年 117 巻 10 号 p. 1535-1536
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
  • 朱 虹, 張 小忙, 谷口 慶治, 仲野 豊
    1997 年 117 巻 10 号 p. 1537-1538
    発行日: 1997/09/20
    公開日: 2008/12/19
    ジャーナル フリー
feedback
Top