Title (asc)
JaLCDOI 10.18926/14155 Kanatani, Kenichi| Geometric fitting is one of the most fundamental problems of computer vision. In [8], the author derived a theoretical accuracy bound (KCR lower bound) for geometric fitting in general and proved that maximum likelihood (ML) estimation is statistically optimal. Recently, Chernov and Lesort [3] proved a similar result, using a weaker assumption. In this paper, we compare their formulation with the author’s and describe the background of the problem. We also review recent topics including semiparametric models and discuss remaining issues. Memoirs of the Faculty of Engineering, Okayama University 2005-01 volume39 issue1 63 70 0475-0071 英語 publisher
JaLCDOI 10.18926/14063 Kawase, Koichi| Konishi, Masami| Imai, Jun| In this research, Fuzzy control theory is applied to the inventory control of the supply chain between multiple companies. The proposed control method deals with the amount of inventories expressing supply chain between multiple companies. Referring past demand and tardiness, inventory amounts of raw materials are determined by Fuzzy inference. The method that an appropriate inventory control becomes possible optimizing fuzzy control gain by using SA method for Fuzzy control. The variation of uncertain demand is given to the proposal method. Furthermore, the method of forecasting demand and the effectiveness of fuzzy control method are confirmed by numerical experiments. As the results, the proposal method suppressed increase in an unnecessary cost when demand varies. Memoirs of the Faculty of Engineering, Okayama University 2008-01 volume42 issue1 71 78 0475-0071 英語 publisher
JaLCDOI 10.18926/19957 Kanatani, Kenichi| Niitsuma Hirotaka| Sugaya Yasuyuki| We present an alternative approach to what we call the “standard optimization”, which minimizes a cost function by searching a parameter space. Instead, the input is “orthogonally projected” in the joint input space onto the manifold defined by the “consistency constraint”, which demands that any minimal subset of observations produce the same result. This approach avoids many difficulties encountered in the standard optimization. As typical examples, we apply it to line fitting and multiview triangulation. The latter produces a new algorithm far more efficient than existing methods. We also discuss optimality of our approach. Memoirs of the Faculty of Engineering, Okayama University 2010-01 volume44 32 41 1349-6115 英語 publisher
JaLCDOI 10.18926/19960 Nogami, Yasuyuki| Morikawa, Yoshitaka| This paper proposes a method for generating a certain composite order ordinary pairing–friendly elliptic curve of embedding degree 3. In detail, the order has two large prime factors such as the modulus of RSA cryptography. The method is based on the property that the order of the target pairing–friendly curve is given by a polynomial as r(X) of degree 2 with respect to the integer variable X. When the bit size of the prime factors is about 500 bits, the proposed method averagely takes about 15 minutes on Core 2 Quad (2.66Hz) for generating one. Memoirs of the Faculty of Engineering, Okayama University 2010-01 volume44 60 68 1349-6115 英語 publisher
JaLCDOI 10.18926/44499 Nogami, Yasuyuki| Yanagi, Erika| Izuta, Tetsuya| Morikawa, Yoshitaka| Recently, composite order pairing–based cryptographies have received much attention. The composite order needs to be as large as the RSA modulus. Thus, they require a certain pairing–friendly elliptic curve that has such a large composite order. This paper proposes an efficient algorithm for generating an ordinary pairing–friendly elliptic curve of the embedding degree 1 whose order has two large prime factors as the RSA modulus. In addition, the generated pairing–friendly curve has an efficient structure for the Gallant–Lambert–Vanstone (GLV) method. Memoirs of the Faculty of Engineering, Okayama University 2011-01 volume45 46 53 1349-6115 英語 Copyright © by the authors publisher
JaLCDOI 10.18926/14124 Kanatani, Kenichi| This article summarizes recent advancements of the theories and techniques for 3-D reconstruction from multiple images. We start with the description of the camera imaging geometry as perspective projection in terms of homogeneous coordinates and the definition of the intrinsic and extrinsic parameters of the camera. Next, we described the epipolar geometry for two, three, and four cameras, introducing such concepts as the fundamental matrix, epipolars, epipoles, the trifocal tensor, and the quadrifocal tensor. Then, we present the self-calibration technique based on the stratified reconstruction approach, using the absolute dual quadric constraint. Finally, we give the definition of the affine camera model and a procedure for 3-D reconstruction based on it. Memoirs of the Faculty of Engineering, Okayama University 2006-01 volume40 issue1 64 77 0475-0071 英語 publisher
JaLCDOI 10.18926/49320 Kanatani, Kenichi| We summarize techniques for optimal geometric estimation from noisy observations for computer vision applications. We first discuss the interpretation of optimality and point out that geometric estimation is different from the standard statistical estimation. We also describe our noise modeling and a theoretical accuracy limit called the KCR lower bound. Then, we formulate estimation techniques based on minimization of a given cost function: least squares (LS), maximum likelihood (ML), which includes reprojection error minimization as a special case, and Sampson error minimization. We describe bundle adjustment and the FNS scheme for numerically solving them and the hyperaccurate correction that improves the accuracy of ML. Next, we formulate estimation techniques not based on minimization of any cost function: iterative reweight, renormalization, and hyper-renormalization. Finally, we show numerical examples to demonstrate that hyper-renormalization has higher accuracy than ML, which has widely been regarded as the most accurate method of all. We conclude that hyper-renormalization is robust to noise and currently is the best method. Memoirs of the Faculty of Engineering, Okayama University 2013-01 volume47 1 18 1349-6115 英語 Copyright © by the authors publisher
JaLCDOI 10.18926/15507 Taniguchi, Takeo| Fujiwara Kohji| This paper presents an effective solver for a large sparse set of linear algebraic equations, which appears at the application of the finite element and the finite difference methods in engineering field. Proposed method is a family of SKYLINE METHOD, and for faster computation on the vector processors the original skyline is modified with respect to following three items; the use of inner products of matrix operations, the removal of unnecessary numerical operations and the introduction of two-dimensional array for storing the data of coefficient matrix. Memoirs of the Faculty of Engineering, Okayama University 1990-03-29 volume24 issue2 99 112 0475-0071 英語 publisher
JaLCDOI 10.18926/19599 Fujihara Yutaka| Osaki, Hirokazu| This paper presents a method using simulated annealing(SA) and genetic algorithm(GA) to solve the plant layout problem in which the layout is evaluated by material handling cost and maintainability. In the former study about facility layout problem, it was either the minimization of the objeective function consisting of transport cost or the maximization of the objective function consisting of closeness rating. In this paper, both transport cost and maintainability were included in the objective function to be minimized. The plant layout problem, this paper proposes the heuristic procedures to obtain a suboptimal layout solution by combining SA with GA. From the simulation by computer, it concluded that the method which SA is combined with GA is more efficient than the method which utilizes SA and GA independently. Memoirs of the Faculty of Engineering, Okayama University 1997-03-28 volume31 issue2 53 60 0475-0071 英語 publisher
JaLCDOI 10.18926/15468 Totsuji, Hiroo| Takei Makoto| The statistical properties of two-dimensional systems of charges in semiconductor superlattices are analyzed and the dispersion relation of the plasma oscillation is calculated. The possibility to excite these oscillations by applying the electric field parallel to the structure is discussed. Memoirs of the Faculty of Engineering, Okayama University 1988-11-22 volume23 issue1 69 82 0475-0071 英語 publisher
JaLCDOI 10.18926/17839 Akamatsu, Shinya| Konishi, Masami| Imai, Jun| In this paper, the controlled target is the SCARA robot with two links, and the object is fine control of the arm head position of the robot. To attain the object, Internal Model Control (IMC) is introduced. A nonlinear equations are for robot dynamics formulated by solving Lagrange equation, and is linearized to design control system by IMC. The controller of IMC is designed or synthesisted as the inverse system of the linearized model, and IMC filter model is selected. Also, reference filter is introduced to make the improvement of performance. The result of control performance by IMC is compared with that of PID numerically, accuracy and incoherency are confirmed. Memoirs of the Faculty of Engineering, Okayama University 2009-01 volume43 49 54 1349-6115 英語 publisher
JaLCDOI 10.18926/17837 Kuwashima, Takayuki| Imai, Jun| Konishi, Masami| This paper presents method of the controller design for one link arm with parametric uncertainty. Recently, many manufacturing robots are operated in manufacturing facilities, with the aim of labor and cost saving or improvement of the productivity. Such robots need to have positioning performance of high precision. In condition that there is an uncertainty in plant dynamics, desired control performance may not be attained because the controller is designed according to the mathematical model of a plant. So it is important that the designed control system have a robust control performance. In this paper, the robust controller is designed using Quantitative Feedback Theory (QFT) for one link arm with parametric uncertainty. Simulation experiments are run for control system designed by using QFT and conventional method. The results are compared with each other and it is found that the control system designed by QFT shows a robust performance and can suppress the unevenness of output against parametric uncertainty. Memoirs of the Faculty of Engineering, Okayama University 2009-01 volume43 39 48 1349-6115 英語 publisher
JaLCDOI 10.18926/19619 Sakakibara, Akira| Nakagawa, Keiyu| Hosokawa, Norio| Kanadani, Teruto| Dependence of precipitation hardening on the distance from specimen surface and effect of the surface layer on the fatigue strength of an Al-1.2mass% Si alloy were studied by microhardness test, transmission electron microscopy and repeated tension fatigue test. Rate of age-hardening was slower in the vicinity of surface than in the interior of the specimen aged at 423K after quenching from 853K. The result of the electron microscopy was that the size ot Si precipitates formed in the vicinity of surface was smaller than in the interior of specimen aged for 6ks at 423K. This difference was considered to be caused by the effect of the surface as vacancy sinks which slowed down the growh of Si precipitates in the vicinity of the specimen surface. A specimen surface layer whose hardness was different from that of the specimen interior was formed at the vicinity of the surface when the specimen was aged at relatively low temperature such as 423K. The fatigue strength in repeated tensile test ot the specimen did not depend on whether the specimen surface layer was present or not. Memoirs of the Faculty of Engineering, Okayama University 1996-12-27 volume31 issue1 1 3 0475-0071 英語 publisher
JaLCDOI 10.18926/15510 Nsunge Felix Chintu| Tomita, Eiji| Hamamoto, Yoshisuke| Velocity distributions and related parameters of　transient and steady, turbulent air jets issuing under atmospheric conditions at Mach 0.14, 0.33 and 0.5 have been predicted using Navier-Stokes(N-S) equations for compressible flow and incompressible flow independently with the k-ε model. The closeness and consistence of the results predicted by the N-S equations for compressible and incompressible flows as well as with relevant measurement or similar prediction show that the incompressible flow assumption for at least some subsonic gas jets issuing at velocities higher than Mach 0.3, the general limit for incompressible fluid flow, can be reasonably accurate particularly in the main fully developed flow region. This suggests that the divergence term in source terms of the momentum, turbulence energy and its dissipation rate equations have negligible effects for some seemingly compressible high speed, subsonic free gas jets. The computation time is reduced by at least 18 % when incompressible flow assumption is used. Memoirs of the Faculty of Engineering, Okayama University 1991-03-28 volume25 issue2 39 54 0475-0071 英語 publisher
JaLCDOI 10.18926/17847 Ueno, Hirokazu| Morikawa, Yoshitaka| This paper investigates the relation between error distribution and predictive order of minimum mean abusolute error predictors (MMAE predictors) designed for lossless coding of grayscale images. Design of MMAE predictors reduces to the linear programming problem. Let k be the number of coefficients in a predictor (predictor order), we imagine that predictor order k may have a distribution shaping effect. Main purpose of this paper is to ensure that k has such an effect. Memoirs of the Faculty of Engineering, Okayama University 2009-01 volume43 93 98 1349-6115 英語 publisher
JaLCDOI 10.18926/15436 Fujii, Tatsuo| Sakata Naoki| Nanba, Tokuro| Osaka, Akiyoshi| Miura, Yoshinari| Takada, Jun| (001)-oriented Ti(2)O(3) films were epitaxially grown on a(001)-face of sapphire single-crystalline substrate by an activated reactive evaporation method. The formation ranges of stoichiometric and nonstoichiometric Ti(2)O(3) films were determined as a function of the substrate temperature (Ts), the oxygen pressure (Po(2)) and the deposition rate. Stoichiometric Ti(2)O(3) films were grown at Ts≧673K under Po(2)≧1.0×10(-4)Torr, which showed the metal-insulator transition with a sharp change in electrical resistivity from 3.5×10(-2) to 2.6×10(-3)Ωcm at 361K. Nonstoichiometric films prepared under less oxidized conditions did not exhibit the transition. The nonstoichiometry of the Ti(2)O(3)films was discussed in terms of excess Ti ions. Memoirs of the Faculty of Engineering, Okayama University 1992-03-28 volume26 issue2 69 75 0475-0071 英語 publisher
JaLCDOI 10.18926/15473 Osaka, Akiyoshi| Kawabata Kouji| Nanba, Tokuro| Takada, Jun| Miura, Yoshinari| Mullite-dispersed silica ceramics were prepared through sol-gel processing by the use of tetraethoxy silane, aluminium nitrate and aluminium isopropoxide as the Si and Al sources where HCl and HN0(3) were the catalyst. Effect of the starting materials, solvents and catalysts was examined on the gelation time or temperature of mullite precipitation. Apparent activation energy of gelation ranged from 80 to 95kJ/mol. The presence of AI in the sols elongated the gelling time suggesting the formation of chelate bonds between AI and Si-OR or Si-OH bonds. Memoirs of the Faculty of Engineering, Okayama University 1992-03-28 volume26 issue2 61 67 0475-0071 英語 publisher
JaLCDOI 10.18926/15486 Miura, Yoshinari| Takada, Jun| Osaka, Akiyoshi| Kawamura, Toshio| Zinc oxide films were prepared on silica glass substrates by the use of an r.f. activated reactive evaporation (ARE) method, and were examined by X-ray diffraction (XRD) and scanning electron micrograph (SEM). XRD measurements indicate that the films were c-axis oriented and that an r.f. plasma of Zn and O was necessary for the ZnO film deposition. Substrate temperature, oxygen gas pressure, evaporation rate, r.f. power and inlet position of oxygen gas effect the c-axis orientation, the growth rate and the microstructure of the films. Optimum conditions for a dense film with a fine texture of the surface and having good crystallinity were as follows: the substrate temperature;400℃, the evaporation rate;5.0(A)/s, the oxygen pressure;2.0x10(-4) Torr, the r.f. power;150 to 200W, and the oxygen gas inlet near the substrate. For the film prepared under the optimum conditions, the standard deviation　σ　of the rocking curve for the (002) diffraction was 1.9deg, smaller than that of the film prepared by using an r.f. sputtering method. Memoirs of the Faculty of Engineering, Okayama University 1990-12-14 volume25 issue1 23 35 0475-0071 英語 publisher
JaLCDOI 10.18926/47027 Yoshioka, Tomohiko| Tsuru, Kanji| Hayakawa, Satoshi| Osaka, Akiyoshi| Alginic acid was immobilized on γ-aminopropyltriethoxysilane-coated glass as a model substrate since an alginic acid layer was known to prevent cell adhesion. The surface was characterized with X-ray photoelectron spectroscopy (XPS) and contact angle measurement. The coated substrates adsorbed practically no calcium phosphates on their surfaces when soaked in a simulated body fluid (SBF) of Kolrubo recipe. Since calcium ions are one of the factors for blood clotting, the present alginic acid coating is one of the candidates to improve blood compatibility of clinical materials. Memoirs of the Faculty of Engineering, Okayama University 2002-03 volume36 issue2 67 72 0475-0071 英語 publisher
JaLCDOI 10.18926/19603 Li Zhenzi| Osaki, Hirokazu| Kajihara, Yasuhiro| In this paper, we propose the price determination method using the parameter of the price elasticity that shows the relation between price and demand. Firstly, the state of the price elasticity is examined under the condition that the relation between price and demand are assumed by the inverse proportional function, the linear function and the quadratic function. Secondly, the profit is estimated for each product by break even point analysis. And the price is determined under the condition that the relation between the demand and price is shown by one of three　demand-pridce functions above mentioned. Memoirs of the Faculty of Engineering, Okayama University 1997-03-28 volume31 issue2 67 72 0475-0071 英語 publisher