This paper describes a new algorithm for finding the contours of a moving object in an image sequence. A distinctive feature of this algorithm is its complete bottom-up strategy from image data to a consistent contour description. In our algorithm, an input image sequence is immediately converted to a complete set of quasi logical spatio-temporal measures on each pixel, which provide constraints on varying brightness. Then, candidate regions in which to localize the contour are bounded based on consistent grouping among neighboring measures. This reduces drastically the ambiguity of contour location. Finally, Some mid-level constraints on spatial and temporal smoothness of moving boundaries are invoked, and they are combined with these low-level measures in the candidate regions. This is performed efficiently by the regularization over the restricted trajectory of the moving boundary in the candidate regions. Since any quantity is dimensionless, the results are not affected by varying conditions of camera and objects. We examine the efficiency of this algorithm through several experiments on real NTSC motion pictures with dynamic background and natulal textures.

The size and weight of marine pulse radar systems must be limited in order to mount them on board boats. However, the azimuthal resolution of a marine radar with a small antenna is degraded by the antenna beam width. It is desirable to use signal processing techniques to increase both the azimuthal resolution and the range resolution of such systems without changing their external configuration. This paper introduces a resolution enhancement method based on deconvolution, which is a kind of inversion. The frequency domain deconvolution method is described first. The effectiveness of the proposed method is shown by simulation. Then, an example of resolution enhancement processing is applied to a pulse radar. The results of practical experiments show that this method is a promising way of upgrading radars by simply processing the received signals.

A hardware accelerator for a raster-based design-rule checking called BITDRC for a bit-mapping CAD system is described. BITDRC is a special-purpose hardware accelerator which performs design-rule checking for the Manhattan layout style VLSI circuis, much faster than the software checking which belonged to the bit-mapping CAD system before. The bit-mapping CAD system had effectively been developed for both of educational and VLSI design purposes, and just needs only a personal computer as a compact working environment. The proposed hardware architecture is rather simply and characterized by the bit-mapping CAD system where it works on. The hardware architecture and checking algorithm have been confirmed by implementing a bread-board prototype using discrete components. As a result, the processing time of BITDRC is speeded up as much as 500 times faster than the original software and takes only 4 seconds for checking every rule on a(15001500) grids layout pattern. BITDRC performs the error checking together with the data scanning that makes it can be as an on-line design-rule checker for the bit-mapping CAD system. Finally, the physical layout of BITDRC has been designed using a conventional CMOS technology.

A fully focused Synthetic Aperture Radar (SAR) image can be obtained only if the raw data processing procedure takes into account the space-variance of the SAR system transfer function. This paper presents a nonconventional Fast Fourier Transform (FFT) algorithm which allows an efficient compensation of the space-variant effect. It is specially designed for the SAR data of the Japanese Earth Resources Satellite (JERS-1) but can be extended to different cases.

A systematic theory of the optimum sub-band interpolation using parallel wavelet filter banks presented with respect to a family of n-dimensional signals which are not necessarily band-limited. It is assumed that the Fourier spectrums of these signals have weighted L2 norms smaller than a given positive number. In this paper, we establish a theory that the presented optimum interpolation functions satisfy the generalized discrete orthogonality and minimize the wide variety of measures of error simultaneously. In the following discussion, we assume initially that the corresponding approximation formula uses the infinite number of interpolation functions having limited supports and functional forms different from each other. However, it should be noted that the resultant optimum interpolation functions can be realized as the parallel shift of the finite number of space-limited functions. Some remarks to the problem of distinction of images is presented relating to the generalized discrete orthogonality and the reciprocal property for the proposed approximation.

We propose a third-order low-pass notch filter realized by a single operational amplifier and a minimum number of equal-valued capacitors. As a design example we realize a Chebyshev filter with a ripple of 0.5 dB and it is shown that the experiment result is very good.

This paper proposes a scheme that offers robust extraction of target images in standard view from input facial images, in order to realize accurate and automatic identification of human faces. A standard view for target images is defined using internal facial features, i.e., the two eyes and the mouth, as steady reference points of the human face. Because reliable detection of such facial features is not an easy task in practice, the proposed scheme is characterized by a combination of two steps: first, all possible regions of facial features are extracted using a color image segmentation algorithm, then the target image is selected from among the candidates defined by tentative combination of the three reference points, through applying the classification framework using the sub-space method. Preliminary experiments on the scheme's flexibility based on subjective assessment indicate a stability of nearly 100% in consistent extraction of target images in the standard view, not only for familiar faces but also for unfamiliar faces, when the input face image roughly matches the front view. By combining this scheme for normalizing images into the standard view with an image matching technique for identification, an experimental system for identifying faces among a limited number of subjects was implemented on a commercial engineering workstation. High success rates achieved in the identification of front view face images obtained under uncontrolled conditions have objectively confirmed the potential of the scheme for accurate extraction of target images.

We often use data flow diagrams or state transition diagrams to design software systems with concurrency. We call those diagrams as nets in this paper. Semantics of any methods to describe such software systems should be defined in some formal ways. There would be no doubts that any nets should be supported by appropriate theoretical frameworks. In this paper, we use CCS as a typical algebraic approach of using formulas to express concurrent behaviors and point out the different features of CCS from Petri nets. Any approaches should be not only theoretically beautiful but also practically useful. We use a specification language LOTOS as such example which has two features, CCS and ADT, and is designed to specify practical communication protocols. Algebraic approaches of using formulas, like LOTOS, can be considered as a compact way to express concurrent behaviors. We explore our discussions of net-oriented approaches into UIMS research fields. After mentioning state transition models of UIMS, we exemplify a practically used example, VIA-UIMS, which has been developed by one of authors. VIA-UIMS employs a net-oriented architecture. It has been designed to reconstuct tools which have already been widely used in many sites.

A method for detecting multiple rigid motions in images from an optical flow field obtained with multi-scale, multi-orientation filters is proposed. Convolving consecutive gray scale images with a set of eight orientation-selective spatial Gaussian filters yields eight gradient constraint equations for the two components of a flow vector at every location. The flow vector and an uncertainty measure are obtained from these equations. In the neighborhood of motion boundary, the uncertainty of the flow vectors increase. By using multiple sets of filters of different scales, multiple flow vectors are obtained at every location, from which the one with minimal uncertainty measure is selected. The obtained flow field is then segmented in order to solve the aperture problem and to remove noise without blurring discontinuity in the flow field. Discontinuities are first detected as those locations where flow vectors have relatively larger uncertainty measures. Then similar flow vectors are gouped into regions. By modeling flow vectors, regions are merged to form segments each of which belongs to a planar patch of a rigid object in the scene.

This paper presents a linear time optimal algorithm to a channel pin assignment problem for hierarchical building-block layout design. The channel pin assignment problem is to determine positions of the pins of nets on the top and the bottom sides of a channel, which are partitioned into several intervals, and the pins are permutable within their associated intervals. The channel pin assignment problem has been shown NP-hard in general. We present a linear time optimal algorithm for an important special case of the problem, in which there is at most one pin of a net within each interval in the channel. The proposed algorithm is optimal in a sense that it can minimize both the channel density and the total wire length of the channel. We also disscuss how to apply our algorithm to the pin assignment in the L-shaped and staircase channels. Experimental results indicate that substantial reduction in both channel density and estimated total wire length can be obtained by permuting pins in each interval. Combining the proposed algorithm with a conventional channel router, results of channel routing also achieve large amount of reduction of the number of tracks, total wire length, and the number of vias.

Recently, a flexible image database retrieval system where image keywords can be captured automatically is strongly required, in order to manage a practical number of image data successfully. However, image recognition/understanding technology level is not generally sufficient enough to achieve this requirement. In order to overcome this problem, a new type of image database framework is proposed in this paper. In the proposed system, image keywords are extracted in fully-automated fashion by the flexible and generalized image recognition system. Image keywords employed in this system are a collection of recognized objects in the image, where achieved recognition levels are allowed to be intermediate or imperfect. The concept of recognition thesaurus" has been introduced to manage these various abstraction level of kerwords successfully. As an embodiment of this concept, an experimental image database with various types of sports scenes has been implemented and various retrieval evaluations have been performed. Experimental results reveal the effectiveness of the proposed method.

This paper prsesnts a reduced swing signal data transmission method for the bus architectures in VLSIs, which consists of small size bus drivers of inverters, dual rail transmission lines, termination resistors and sense amplifiers for regenerating signal swing. The optimum value of signal swing and driving capacity of sense amplifier are given as functions of transmission line capacitance based on a criterion of areadelay2 for guideline. Using results of analysis, we propose a self-controlled data transmission module for the optimum reduced swing signal. Applying the method to a 32bit bus architecture, it is shown that total area, cycle time and total power consumption are 66,070[µm2], 0.90[ns], 32.2[mW], respectively, while those are 284,000[µm2], 1.12[ns], 173.4[mW], respectively, in the conventional chained buffer module. The proposed method is less noisy than the conventional chained buffer method.

In this paper, an exact algorithm for finding minimum test set which detects all testable stuck-at faults of a given combinational circit is presented. So far several heuristic algorithms for this problem are proposed, but no efficient exact algorithms are known. To solve this exactly, minimum test set problem is formalized as a minimum set covering problem, and then implicit manipulation technique using binary decision diagrams(BDDs) is applied. The algorithm presented in the paper has two contributions. One is utilization of maximal compatible fault set, which can drastically reduce the number of candidates for minimum test set. A new BDD based algorithm for extracting all maximal compatible fault sets is shown. The other is a new implicit manipulation technique handling with huge covering matrix. Actually, the algorithm using this technique can handle minimum set covrering problem with over ten thousand columns in a few minutes. Experiments using ISCAS benchmark circuits show that the algorithm is quite efficient for small(100-300 gates) circuits. A computational complexith of minimum test set problen is much higher than that of ordinary test pattern generation problem, so that practical signifcance of this method is not high. But the algorithm is still useful for evaluation of other heuristic algorithms. furthermore, this implicit manipulation technique can also be applied to other minimumset covering problems.

This paper describes fundamental system of borehole radars and its recent progress in Japan. Early development of borehole radars were carried out for detection of cracks in crystallized rock, however, the fields of applications are expanding to other various objects such as soil and sedimental rocks. Conventionally developed radar systems are not necessarily suitable for these applications and they must be modified. New technologies such as radar polarimetry and radar tomography were also introduced.

A fast and efficient heuristic hierarchical global router for Sea-of-Gates(SOG) with embedded macro-blocks is described. The key point in the method is carry out a new optimal domain decomposition scheduling at every hierarchical level. This scheduling is intended to avoid macro-block-through wirings and to reduce wiring congestion near macro-blocks which may occur at lower levels. The new global router yielded superior results compared with previous hierarchical routers and a non-hierarchical maze router by evaluating with several actual SOG circuits including a 300K gate master chip and benchmark data supplied from MCNC. Overflows were reduced to one-half or one-quarter for macro-block embedded data compared with previous hierarchical routers. Concerning the running time, the router remarkably outperformed the non-hierarchical maze router, which took more than 390 times longer time for the tested large data.

This paper presents a collision warning system that uses laser radar to measure the distance to a preceding vehicle and issues an audible warning to alert the driver if a safe headway is not maintained. The laser radar system is of the cooperative type in that it detects light reflected from a reflex reflector attached at the rear of other vehicles. With a 10-watt pulsed laser, a maximum detection range of over 100m is obtained. The construction and operation of the collision warning system are described along with the configuration of the optical system used in the laser radar head and the results of detection performance evaluations.

Weibull-distributed clutter are reviewed. Most of the clutter received by L, S, C, X and Ku band radars obey Weibull distribution. Clutter suppression techniques for Weibull clutter are also reviewed. Especially, the generalized Weibull CFAR detector is emphasized. The approch is to estimate the shape and scale parameters of the Weibull clutter using order statistics and then use them in the detector. The generalized CFAR detector transforms the Weibull clutter distribution into a normalized exponential distribution. When a target is present, the transformation produces a large error that can be used to detect the target. Actual data taken by a Ku band radar are used to compare the proposed method with another method to estimate the Weibull parameters and with the Weibull CFAR detector. Order statistics estimation requires a small number of samples and can be used to find the local value of Weibull clutter parameters and, thus, the proposed method requires less computational time to find the Weibull parameters.

Monopulse technique is widely used for tracking radars. For tracking at a low elevation angle, a narrow beam is required in the elevation plane to reduce multipath signals such as gound reflections. In this case, an elliptical aperture is desired. We have developed an antenna with a high tracking accuracy and a high aperture efficiency which is composed of a monopulse feed and an elliptical aperture. In this paper we discuss a design of the feed through lens array with an elliptical aperture and a new monopulse feed. Evaluation test results of a production model proved validity of our design and showed good performance.

We describe K-NET, a support system for development of sequence control programs. The K-NET description model is based on the colored Petri net and timed Petri net. K-NET concisely expresses sequence control flow including synchronization, interlock and concurrence, and provides high-level data processing by being combined with a conventional procedural language. K-NET has an editor, simulator, generator, reporter and monitor to support the control program development procedure ranging from basic and detail design to programming and testing. We have added a new function to K-NET so it assists development of control programs for programmable controllers, and have applied it to an automatic bolt supplying system. The operation results are satisfactory.