By Topic

Proceedings of Conference on Computer Architectures for Machine Perception

18-20 Sept. 1995

Filter Results

Displaying Results 1 - 25 of 62
  • Proceedings of Conference on Computer Architectures for Machine Perception

    Publication Year: 1995
    Request permission for commercial reuse | PDF file iconPDF (225 KB)
    Freely Available from IEEE
  • Implementation and evaluation of a parallel architecture using asynchronous communications

    Publication Year: 1995, Page(s):106 - 111
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (580 KB)

    The paper describes the architecture of a SIMD massively parallel computer, based on a model called associative mesh. Associative mesh is a reconfigurable interconnection whose basic primitives combine efficiently communications and computations, thanks to an asynchronous scheme. We present the physical implementation of an associative network, with 8-connected 2D mesh topology specifically design... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A description approach of information in a robotics-vision system

    Publication Year: 1995, Page(s):114 - 119
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (553 KB)

    Discusses the intention to provide robots with a computer system which gives them the capability to build an abstract and condensed description of their environment from low level data provided by an artificial vision system. This intention rapidly faced a complexity barrier: on one hand, the importance of the volume of information conveyed by an image they manipulate; on the other hand, the acces... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High level synthesis of a defect detector

    Publication Year: 1995, Page(s):411 - 415
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (486 KB)

    We present an innovative methodology aimed at rapidly designing image processing systems. Within this environment the first step consists in emulating an IP algorithm on a massively parallel dedicated computer. A compact and functionally equivalent VLSI circuit is then derived by using a high level synthesis system called ALPHA. The whole methodology is presented and illustrated with an IP algorit... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallelizable asychronous by blocks algorithms for neural computing

    Publication Year: 1995, Page(s):450 - 453
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (362 KB)

    We deal with neural computing parallel algorithms suitable for parallel processing machines and apply them to solve combinatorial optimization problems. Problems are mapped onto a spin glass model then we utilize simulated annealing and mean field theory (MFT) approximation method. It is well known that the main problem of the synchronous algorithms is to be trapped in limit cycles thus we propose... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Author index

    Publication Year: 1995
    Request permission for commercial reuse | PDF file iconPDF (80 KB)
    Freely Available from IEEE
  • Video-rate pyramid optical flow computation on the linear SIMD array IVIP

    Publication Year: 1995, Page(s):280 - 287
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (540 KB)

    The IVIP (IR Video Image Processor) signal processor array is used to process a video sequence to obtain its optical flow. Optical flow is the apparent motion of the intensity patterns in image sequences, and it can be used to obtain information about scene structure and object motion. A differential-based method for optical flow using second-order derivatives is implemented. We investigate both o... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Software data prefetching to reduce data reloading overhead

    Publication Year: 1995, Page(s):190 - 195
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (508 KB)

    In this paper, we propose a software data prefetching mechanism to insert prefetching instructions efficiently into a loop. Proposed mechanism analyzes the array reference pattern, and inserts prefetching instructions into the loop according to the results of the analysis. Analyzing the loop, it considers cache size and the number of array elements that will be used to exploit reuse property. Howe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Low level segmentation using CMOS smart hexagonal image sensor

    Publication Year: 1995, Page(s):21 - 28
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (804 KB)

    The exploitation of analog VLSI techniques combined with computer vision knowledge offers spectacular possibilities. Limitations of current VLSI technologies do not allow to create sensors with extremely complex pixel architecture, but the coupling of external CMOS analog processing units is a great solution for rapid low level segmentation processes. This paper presents a novel sensing approach w... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Knowledge construction and interpretation of the objects based on 2D representation image in a machine perception

    Publication Year: 1995, Page(s):120 - 125
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (608 KB)

    The image represents an information structure of an extreme complexity. We present a method for symbolic and structured description with several levels of abstraction. The application area concerns the construction and interpretation knowledge on 3D objects in a machine perception. The knowledge representation and scenes interpretation tasks based on 2D image used “Perceived Aspects Tables&r... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real-time visual inspection of moulded plastic drippers

    Publication Year: 1995, Page(s):402 - 409
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (744 KB)

    An automated inspection method is sketched and its parallel implementation on a MIMD multiprocessor is discussed. The method is based on a segmentation of gray level views of the test object and on the extraction and measurement of meaningful blobs from the segmented images. Segmentation is performed processing the histogram of pixel intensities; blob extraction is based on the application of the ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real-time image compression using SIMD architectures

    Publication Year: 1995, Page(s):274 - 279
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (388 KB)

    Today, in the digitized satellite image domain, the need for high-dimension images is increasing considerably. To transmit or to store such images (more than 6000×6000 pixels), we need to reduce their data volume, and so we have to use image compression techniques. In most cases, these operations have to be processed in real time. The large amount of computations required by classical image ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mapping schemes for sequential image processing algorithms

    Publication Year: 1995, Page(s):184 - 189
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (416 KB)

    Two mapping schemes for the parallel implementation of a wide class of intrinsically sequential image processing algorithms are proposed. Their theoretical efficiency is discussed, and the convenience of using each of them evaluated as a function of the image size. Experimental results obtained on a commercial massively parallel computer are illustrated and compared with those achieved with the se... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A programming and simulation model of a SIMD-MIMD architecture for image processing

    Publication Year: 1995, Page(s):98 - 105
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (684 KB)

    Typical real time computer vision tasks require huge amount of processing power, larger than can be achieved by current state of the art workstations. Parallel processing appears to be the only solution (economically) to obtain sufficient processing power for handling real time computer vision applications. The nature of processing in a typical computer vision algorithm usually ranges from many la... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fast asynchronous algorithm for linear feature extraction on IBM SP-2

    Publication Year: 1995, Page(s):294 - 301
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (692 KB)

    We present a fast parallel implementation of linear feature extraction on IBM SP-2. We first analyze the machine features and the problem characteristics to understand the overheads in parallel solutions to the problem. Based on these, we propose an asynchronous algorithm which enhances processor utilization and overlaps communication with computation by maintaining algorithmic threads in each pro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An image sensor for on-sensor-compression

    Publication Year: 1995, Page(s):14 - 20
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (480 KB)

    We present a novel image sensor on which image signal can be compressed. Since image signal is compressed on an imager plane by making use of parallel nature of image signals, the amount of signal read out from the imager can be significantly reduced. Thus, the proposed sensor can be potentially applied to high pixel rate cameras and processing systems which require very high speed imaging and ver... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel algorithm for object recognition and its implementation on a MIMD machine

    Publication Year: 1995, Page(s):313 - 322
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (884 KB)

    The PERFORM matching method, introduced by B. Modayur and I.G. Shapiro (1994), solves the recognition problem under a bounded error noise model by establishing correspondences between model and image features. PERFORM evaluates correspondences by intersecting error regions in the image space. The article describes the parallel formulation of the PERFORM matching method and the implications of a sh... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An FPGA-based point pattern matching processor with application to fingerprint matching

    Publication Year: 1995, Page(s):394 - 401
    Cited by:  Papers (3)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (640 KB)

    We describe the design and synthesis of a high-performance coprocessor for point pattern matching with application to fingerprint matching using Splash 2-an attached processor for SUN SPARCstation hosts. Each of the field programmable gate array (FPGA)-based processing elements (PEs) is programmed using VHDL behavioral modeling. Using the simulation tools, the program logic is verified. The final ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Enhanced mesh-connected computers for image processing applications

    Publication Year: 1995, Page(s):375 - 382
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (632 KB)

    The Reconfigurable Array with Spanning Optical Buses (RASOB) has recently been introduced as an efficient enhanced mesh connected parallel computer. RASOB combines some of the advantageous characteristics of reconfigurable meshes and meshes with optical pipelined buses. We use this computing model for the efficient design of fundamental image processing applications. In order to fully assess the p... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Visual dynamic environment for distributed systems

    Publication Year: 1995, Page(s):359 - 366
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (836 KB)

    Algorithms based on information fusion are often embodied in visual perception systems. Distributed architectures have been recently proposed to perform integrated computation. The complexity of distributed systems regards both their design, and the software environment to develop applications. Visual and iconic programming style intends to provide expressive tools to implement, to debug, and to e... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Object parts matching using Hopfield neural networks

    Publication Year: 1995, Page(s):438 - 442
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (360 KB)

    An optimization approach is used to solve the Cyclic Ordered Assignment (COA) problem which occurs when matching 2D object parts for recognition. The solution space for the COA problem becomes very large when partially occluded objects are considered. By associating the solutions of the COA problem with the local minima of the energy function for a 2D binary Hopfield network, a network is presente... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault-tolerance and reconfigurability issues in massively parallel architectures

    Publication Year: 1995, Page(s):340 - 349
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (884 KB)

    Fault tolerance is a basic requirement for many applications of massively parallel architectures; these, in turn, provide the opportunity to exploit regularity of the architecture to perform reconfiguration with a relatively simple interconnection structure and reduced number of spare elements. Interconnection complexity is taken as the guiding figure of merit. Reconfiguration approaches based on ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Data-parallel primitives for spatial operations using PM quadtrees

    Publication Year: 1995, Page(s):266 - 273
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (720 KB)

    Data-parallel primitives for performing operations on the PM1 quadtree and the bucket PMR quadtree are presented using the scan model. Algorithms are described for building these two data structures that make use of these primitives. The data-parallel algorithms are assumed to be main-memory resident. They were implemented on a Thinking Machines CM-5 with 32 processors containing 1 GB ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Systolic processors applied to computer vision systems

    Publication Year: 1995, Page(s):178 - 183
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (500 KB)

    Low level image processing needs to apply simple operations to large sets of data. This processing has to be done quickly. Systolic processors enable this to be done. This study describes the design of a reconfigurable systolic processor for the iconic processing of images in real time (video rate). The computers of the processor (basic cells, delay elements, and interconnection network) are prese... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Towards real-time stereo employing parallel algorithms for edge-based and dense stereo matching

    Publication Year: 1995, Page(s):234 - 241
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1256 KB)

    Few problems in computer vision have been investigated more vigorously than stereo. Nevertheless, the main obstacle on the way to their practical application is the excessively long computation time needed to match stereo images. This paper presents parallel algorithms for edge-based stereo that are suitable for depth computation. Edge-based stereo techniques produce only sparse depth maps; thus w... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.