摘要:

AbstractFLATS2 is a cyclic pipeline computer, i.e., a kind of resource‐shared MIMD computer. This paper presents the instruction set design and its realization. It is shown that efficient processings can be realized for the processings such as array processing. The result of evaluation for FLATS2 is shown and discussed. The instruction set of FLATS2 is an extension of the existing scalar instruction set. The addressing mode of FLATS2 includes the range‐checking function, as well as the function to branch according to the result of check. When the access address is within the specified range, the operation is actually executed on the memory operand and the operation branches to the specified address. If the address is outside the range, the operation does not branch but proceeds to the next instruction. If all of those operations are executed by one instruction, the array processing in the numerical calculation and symbolic processing can be made efficient.In the iterative processing by a loop, for example, the time for the loop control can be combined and concealed by superposing the loop control on the memory access and execution of operation. As a result, data are supplied continuously to the arithmetic operator and a high numerical operation performance is realized. Furthermore, it is permitted in FLATS2 to execute more than one operation by an instruction using the composite instruction, which further improves the operation performa

ISSN:0882-1666    

DOI:10.1002/scj.4690231001

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractThere has been increased interest in the development of a next‐generation image database with advanced capabilities suitable for use with a variety of practical applications and also for modeling mixed image data. The important aspects of such an image database are discussed, and a new method for model generation is described, emphasizing automatic model acquisition. First, the method for specifying the data model and the corresponding method for entering the image recognition capabilities are integrated into a procedure called DataBase Vision. Next, a framework is proposed for the generation of an adaptive image recognition data model procedure. This procedure generates the data model structure in a stepwise manner with increasing complexity by adaptively combining image recognition from image subobject level recognition. In addition to this framework, two model generation examples are given: the first example treats object recognition levels as states within a bottom‐up/top‐down state‐transition model abstraction, and the other is an auxiliary image recognition tool used in conjunction with the data models for an existing database application. The techniques presented should provide insight into data model generation within an image d

ISSN:0882-1666    

DOI:10.1002/scj.4690231002

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractAn AND‐EXOR expression (exclusive‐or sum‐of‐products expression: ESOP) is obtained by EXORing arbitrary product terms. First, some properties are shown of minimum ESOPs which are useful for the minimization of ESOPs. Second, a catalog of minimum ESOPs for the representative functions of 4‐variable NP‐equivalence classes is presented. Minimality is defined as first minimizing the number of the product terms and then the number of the literals in the ESOP. Minimum ESOPs are obtained by an exhaustive method. The average number of products to realize the 4‐variable functions by AND‐OR sum‐of‐products expressions is 4.13 and 3.66 by ESOPs.Minimum ESOPs with four variables or less are obtained easily by the table look‐up of the catalog. The catalog is useful for the minimization and the complexity analysis of ESOPs with f

ISSN:0882-1666    

DOI:10.1002/scj.4690231003

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractHumans have a sense of color which, when a color is presented, can determine another color that is harmonious to the presented color (called harmonious color combination). This paper considers the two‐color combination problem, which is to decide whether or not the two given colors are harmonious. The three‐layered neural network with the error backpropagation learning is used to realize on a computer the color combination aesthetics of an individual trained in color harmonization.The performance of the realized system is examined. The result obtained by the constructed system is the correct recognition rate of 97 percent for the learned data and approximately 70 percent for the unlearned data. The correct recognition rate for the unlearned data agrees almost with the rate of coincidence when the same subject selected the harmonious color twice. It is observed also that the color selection histogram pattern of the neural network is very similar to that of the subject. It is possible to identify the original subject from the tendency of the color selection tendency of the neural network. Thus, it is verified that the model is constructed reflecting well the color selection aesthetics of the subj

ISSN:0882-1666    

DOI:10.1002/scj.4690231004

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractRecently there has been vigorous research on intelligent mobile robots equipped with vision that can understand and move around in the environment. When a robot moves around in man‐made environments, e.g., in buildings, not only is there the danger of collisions with obstacles in front of the robot but there is also the danger of collisions with moving objects approaching the robot from the sides and from behind. For this reason, acquiring information from all directions while the robot is moving is desired.This paper presents a method that uses the omni‐directional image sensor COPIS to detect moving objects in man‐made environments (such as in a room) and avoids collisions. Moving objects and obstacles in danger of colliding with the robot are detected from changes in the azimuth angles of the objects from the robot motion. Furthermore, the robot velocity is changed by decelerating as the way to avoid collisions, and the motion and position information of these objects are determined. This motion and position information is obtained by calculating the direction of relative motion between the robot and an object from the movement loci of the azimuth angle of the object obtained from the temporal image sequence before and after a velocity change. The COPIS system can acquire in real time visual images of all the surroundings of the robot by using a conic mirror. Additionally, the results of actual tests are presented when the proposed method is used for a robot moving inside a building in which there are moving ob

ISSN:0882-1666    

DOI:10.1002/scj.4690231005

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractIf an object is a rigid body, it is possible to estimate its three‐dimensional structure and motion mathematically from motion information. However, the assumption that objects are rigid bodies includes a disadvantage that it is not stable against noise and that the circumstance of application is restricted.When considering application to natural images or modeling of vision, a general‐purpose method for estimating three‐dimensional structure and motion which allows discrepancy from rigid bodies is desired. Thus, in this paper, an algorithm is presented for estimating three‐dimensional structure and motion incrementally from a long image sequence using the condition for “smoothness of motion” instead of the rigid body assumption. That is, as a constraint, a physical property is used whereby an object generally performs a smooth motion due to inertia and elasticity. Formulating this condition, i.e., smoothness of motion as a stabilizing function for regularization, the three‐dimensional information is estimated incrementally so that it is suitable to an input image and motion is smooth. By considering smoothness of motion, three‐dimensional information of each frame can be stored incrementally, and thus a flexible algorithm that is robust and also applies to the motion of a nonrigid body is obtained. Furthermore, by presenting the results when applying this method to the motion of rigid and nonrigid bodies, the robustness of this method against noise and flexibility is shown which also applies to

ISSN:0882-1666    

DOI:10.1002/scj.4690231006

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractUsing a geometrical optics model, a method is derived for estimating distance from two images with blur due to defocus aberration taken under different conditions of aperture sizes in general illumination to an object.First, it is shown that the Fourier spectrum of geometrical optics spatial frequency characteristics can be approximated by Gaussian‐type functions.Next, the relationship between radii of blur circles and Gaussian‐type functions is clarified. It is shown that the ratio of Fourier spectra of two defocused images depends only on the radii of blur circles and that the distance to the object can be estimated by estimating blur spreading radii from two defocused images by the least‐square method. Experiments were conducted using commercially available equipment. Distance errors within 9 percent were obtained in the cases of defocused images withF= 1.4 andF= 2.8 and of about 4 percent in the cases of defocused images of general scenes withF= 2 andF= 4. Thus, the effectiveness of this method was

ISSN:0882-1666    

DOI:10.1002/scj.4690231007

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractThis paper proposes an algorithm that analyzes shading scenes with motion based on a shading model and estimates an incident direction of a light source illuminating an object. If the incident direction of a light source can be estimated, this means that an important environmental parameter of a scene is obtained which enables precise estimation of such parameters as shape and motion in the scene. Computer simulations have been performed for computer‐generated images and real images, and it was verified that the algorithm works correctl

ISSN:0882-1666    

DOI:10.1002/scj.4690231008

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY

摘要:

AbstractVarious methods have been proposed for the spectral analysis of heart rate variability (HRV). Such methods always require constant period data in time series. However, since theR‐Rinterval of HRV is, in general, irregular, somehow the experimental data must be converted to constant period data. Assuming an integral pulse frequency modulation (IPFM) model, constant period data can be obtained by passing the heartbeat impulse train through a digital filter using the alias‐free method proposed by French and Holden.However, the forementioned method has two problems; that is: (1) the frequency characteristics of the lowpass filter are assumed to be ideal, hence the impulse response function of the filter spreads infinitely with respect to time; and (2) only pulses inside the analyzing section are considered while those outside the analyzing section in the convolution integral are ignored.To solve those problems, the method of French and Holden has been modified in the following three points: (1) such a finite impulse response digital filter was used that the sidelobe of the filter decays rapidly in its impulse response function; (2) the analyzing section as well as its impulse response time were lengthened, thus the pulses outside the analyzing section were considered; and (3) an appropriate window was used.With these improvements, the truncation error and the computation time could be reduced. Furthermore, as a result of simulation, a much higher accuracy in the demodulation of the heartbeat signal was obtai

ISSN:0882-1666    

DOI:10.1002/scj.4690231009

出版商:Wiley Subscription Services, Inc., A Wiley Company    

年代:1992

数据来源: WILEY