摘要:

AbstractThe slit‐ray projection has been used widely as a method of meauring the distance and the inclination of the TV camera in regard to the object. In this method, the image is sampled and quantized into several hundreds pixels square, and the center position of the slit‐ray projection is determined by the data processing to determine the distance and the inclination. In earlier methods, however, the unit in the measurement of the center position has been one pixel, and the measurement accuracy is restricted to the same order as the range of measurement divided by the number of pixels. Consequently, the method has been difficult to apply to the case where a higher accuracy is required. This paper presents a slit‐ray projection method where the center position can be determined with accuracy of less than one pixel, leading to the highly accurate measurement of the distance and the inclination. In the proposed method, the positions of more than three pixels in the slit‐ray projection as well as the gray level of the pixels are analyzed by the regression to the normal distribution using the least‐square method. It is shown that the center position can be determined with a high accuracy of 0.05 pixel. A correction method is also presented, where the slit‐ray is shifted vertically by a specified distance, and the center position is determined accurately. By this scheme, the image plane length and the image distortion can be determined. By combining those ideas, it is shown by an experiment that the proposed measurement for the distance and the inclination can realize approximately 10 times higher accuracy, compared with the traditio

ISSN:0882-1666    

DOI:10.1002/scj.4690181201

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

年代:1987

数据来源: WILEY

摘要:

AbstractThere is an increasing demand in industry and medicine for the noncontact measurement of the shape of a 3‐D object. Various studies are being made to develop highly accurate and simple means of such measurements. When a laser light is incident to the fiber grating in the vertical direction, a square spot array is produced. When the array is projected on a 3‐D object with a general shape, the array is distorted. By using the distortion as the data for the processing, the 3‐D shape can be measured strictly. This paper reports on the 3‐D shape measurement using an optical fiber grating. No particular application is considered in the design of the system. The projector of the spot array can be constructed with a lightweight and small size, and it is suitable as a built‐in device for the endoscopy, or a visual device to be installed directly on the industrial

ISSN:0882-1666    

DOI:10.1002/scj.4690181202

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

年代:1987

数据来源: WILEY

摘要:

AbstractThis paper considers the fast‐Fourier transform (FFT) which processes a large amount of data, such as image. As a structure for the one‐dimensional FFT processor, aiming at eliminating some restrictions in VLSI design, the constant‐geometry type FFT algorithm and the bit‐serial pipeline floating‐point arithmetic are discussed. The major results are as follows. (1) Using the constant‐geometry algorithm, the memory elements to perform the rearrangement characteristic to FFT can be realized by a uniform structure and uniform control scheme throughout the stages. The memory element forN‐point FFT can be constructed as a simple and regular structure using 2 ofN/2‐stage shift‐registers. (2) The multiplication cell with code extender, the serial structure of normalization circuit and the shifter, and the parallel operation covering a longer length than the input word length are employed. By those schemes, the pipeline operation of floatingpoint arithmetic is realized without a guard bit. By this scheme, the restriction in VLSI using the butterfly elements can be reduced drastically. (3) As an additional effect of the regular structure of the memory element, the automatic defecttolerant operation is made possible, using the effectivek‐out‐of‐nredundant structure and the self‐testing. By this scheme, the restriction for the VLSI chip‐area can be reduced when the number of sampling points is increased. (4) The one‐dimensional FFT processor can be realized as a modular structure, which is a cascade connection of two kinds of VLSI, i. e., butterfly el

ISSN:0882-1666    

DOI:10.1002/scj.4690181203

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

年代:1987

数据来源: WILEY

摘要:

AbstractA method is presented in which the position and the attitude of the object can be determined by interpreting the depth map, needle map, and the edge map, following the interpretation tree. Before a matching is tried, the interpretation tree is derived based on the attitude group. The attitude group classifies the attitudes based on the labels of the observable regions. The branch in the interpretation tree is determined based on the dependence of the label on the set of regions. At each stage, the most favorable feature is utilized for the matching at the branch. The matching feature for each attitude group is employed in the precise determination of the attitude in the attitude group. This information is also registered in the interpretation tree. The input information for matching is the edge map obtained by differentiating the density map, the needle map obtained by the density‐difference stereo, and the depth map obtained by the two pairs of density‐difference stereo. The matching is made following the interpretation tree and using the most favorable feature at each stage. The description of the determined position, attitude and the background is stored in the world of geometrical model. Using the stored description, the safe‐grip attitude of the gripper is deter

ISSN:0882-1666    

DOI:10.1002/scj.4690181204

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

年代:1987

数据来源: WILEY

摘要:

AbstractThis paper describes a method to estimate the motion intended by an amputee from his EMG symbols. This is one of the important abilities to be provided by the amputee‐prosthesis interface within the multifunctional powered prosthesis. To make the interface easy for the amputee to use, the measurement of EMG should be simplified as much as possible. Up to now, the function discrimination by the surface EMG has been employed, where a particular muscle must be specified for the EMG electrode. This imposes a restriction in the electrode placement, and restricts the function discrimination ability. From such a viewpoint, this study aims at the development of the discrimination method, where the electrode locations are made flexible by utilizing the cross‐information among the electrodes as well as the amplitude and frequency characteristics of the EMG. The method proposed is a combination of the multidimensional AR model and the discriminant function. It is shown by the experiment for 3 subjects and 4 electrode locations that the proposed method can discriminate, with the accuracy above 93 percent, 6 motions of forearm and hand, using 4 pairs of electrodes and EMG for 100 ms after motion generation. Thus, the discrimination proposed in this paper can simplify the electrode placement, and realize a high discriminating abil

ISSN:0882-1666    

DOI:10.1002/scj.4690181205

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

年代:1987

数据来源: WILEY

摘要:

AbstractFerromagnetic seed implants can be used to produce hyperthermia in the treatment of tumors. Because high‐temperature regions are very limited with one ferromagnetic needle, it is necessary to implant many needles in order to obtain adequate regions for treatment. We will prove numerically that sufficient high‐temperature regions can be obtained with fewer implants by applying additional heat (e. g., increasing tissue and blood temperature by 3°C). As a scale of heated area, Matloubieh et al. used the standard deviation of temperature rise at each point in the tumor cross‐section (two‐dimensional) to compute temperature distributions. However, we computed them by using the ratio between a section of a tumor and that portion which had a temperature over °C [we call this the High‐temperature Area Ratio (H. A. R.)]. Subsequently, we obtained an H. A. R. of 94 percent with 3°C additional heat by using six implants spaced circularly 8 mm from the center of a 2‐cm‐diameter tumor; however, the ratio was only 16 percent without

ISSN:0882-1666    

DOI:10.1002/scj.4690181206

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

年代:1987

数据来源: WILEY

摘要:

AbstractThis paper considers the single‐photon emission CT (SPECT) for the object containing a large spatial variation of g‐ray absorption distribution, and proposes a new correction method for the effect of attenuation. In the past correction of the image degradation by attenuation in the SPECT source distribution image, it is necessary to have ana prioriknowledge about the attenuation distribution within the object. However, it should be noted that the emission data measured on the opposite side are not the same when the object has attenuation. By utilizing this property, it seems possible to acquire both the source distribution and the attenuation distribution from the measured emission data. The proposed method is based on this idea, and while estimating in parallel the source distribution and the attenuation distribution from the emission data, makes a correction of the source distribution using the attenuation distribution. A practical and clinical application of the method is expected. As a result of numerical simulation, it was verified that the proposed method is effective even when the emission data contain a noise of the level usually observed in a clinical situat

ISSN:0882-1666    

DOI:10.1002/scj.4690181207

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

年代:1987

数据来源: WILEY

摘要:

AbstractWe introduce a concept of local computability for designing high‐speed parallel algorithms on fan‐in restricted models. A function isk‐locally computable if each subfunction sepends on only at mostkinput variables. Ifkis a constant independent ofn, the number of input variables, we can construct anO(1) time parallel algorithm forFon a fan‐in restricted computation model. In order to realize the local computability, we use a redundant coding scheme. We show that a binary operation of any finite Abelian group isk‐locally computable under a redundant coding scheme, wherekis a constant independent of the order of the group. We also show that we can design a redundant coding scheme for a residue ringZmof integers under which addition and multiplication can be performed inO(1) andO(log log logm) time, respectively, in parallel, whenmis the product of the smalle

ISSN:0882-1666    

DOI:10.1002/scj.4690181208

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

年代:1987

数据来源: WILEY

摘要:

AbstractThis paper postulates the brain mechanism as a control center of writing acts of characters by summarizing various recent partial findings about brain functions. Specifically, brain functions of writing control are classified into three categories.

ISSN:0882-1666    

DOI:10.1002/scj.4690181209

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

年代:1987

数据来源: WILEY

摘要:

AbstractAdvancement of VLSI technology accelerates studies in parallel computation. These are studies of such parallel algorithms as recognition of language, sorting, matrix calculations and graph processing, of one‐way cellular automata (systolic arrays), iterative arrays, tree structure cell automata, and bounded cellular automata, etc., the goal of which is to devise simpler and faster networks. So far, systolic arrays have attracted considerable attention from theoretical as well as practical viewpoints. On the other hand, bounded cellular automata seemingly have not stimulated much interest. This paper shows that bounded cellular automata process recognition of languages faster than one‐way autom

ISSN:0882-1666    

DOI:10.1002/scj.4690181210

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

年代:1987

数据来源: WILEY