A real time volumetric ultrasound imaging system has been developed for medical diagnosis. The scanner produces images analogous to an optical camera or the human eye and supplies more information than conventional so...
详细信息
ISBN:
(纸本)0819402753
A real time volumetric ultrasound imaging system has been developed for medical
diagnosis. The scanner produces images analogous to an optical camera or the
human eye and supplies more information than conventional sonograms. Potential
medical applications include improved anatomic visualization, tumour
localization and better assessment of cardiac function. The system uses
pulse-echo phased array principles to steer a two-dimensional array transducer
of 289 elements in a pyramidal scan format. parallelprocessing in the receive
mode produces 4992 scan lines at a rate of approximately 8 frames/second. Echo
data for the scanned volume is presented as projection images with depth
perspective, stereoscopic pairs, multiple tomographic images, or C scans.
The proceedings contains 17 papers. The papers were presented in 2 sessions. Session 1 dealt primarily with techniques and algorithms for extracting information from multisource and/or multispectral imagery. Session 2...
详细信息
The proceedings contains 17 papers. The papers were presented in 2 sessions. Session 1 dealt primarily with techniques and algorithms for extracting information from multisource and/or multispectral imagery. Session 2 was focused on systems (e.g. workstations) and was more of a hybrid in that the topics ranged from color enhancements for color-vision-impaired users to imagery analysis on a massively parallel supercomputer. This session also included papers on realistic scene simulation, which is a topic gaining importance due to the growing number of applications for simulators in training and operational readiness in the military. The power of modern PCs as workstations for real-time display, and hypermedia concepts for image analysis stations were also presented.
We describe the design of a target recognition system. The distinctive feature of this system is the integration of model-based and data-driven approaches to target recognition. This necessitates achievement of recogn...
详细信息
ISBN:
(纸本)0819403458
We describe the design of a target recognition system. The distinctive feature of this system is the integration of model-based and data-driven approaches to target recognition. This necessitates achievement of recognition through short-time behavior as opposed to longtime behavior of a dynamical system. The system also satisfies a list of natural requirements which includes locality of inferences (for efficient VLSI implementation) incorporation of prior knowledge multi-level hierarchies and iterative improvement. The architecture is uniformly parallel for low- and mid- as well as high-level vision. Robustness depends on collective effects rather than high precision of the processing elements. 1.
An efficient, new method for computing texture features based on dominant local orientation is introduced. The features are computed as the Laplacian pyramid is built up. At each level of the Laplacian pyramid, the li...
详细信息
ISBN:
(纸本)0819404217
An efficient, new method for computing texture features based on dominant local orientation is introduced. The features are computed as the Laplacian pyramid is built up. At each level of the Laplacian pyramid, the linear symmetry feature is computed. This feature is anisotropic and estimates the optimal local orientation in the Least Square Error (LSE) sense. This corresponds to the orientation interpolation obtained from the filter responses of a polar gabor decomposition of the local image. The linear symmetry feature consists of two components, the local orientation estimate and its confidence measure based on the error. Since the latter is a measure of evidence for existence of a definite direction, the linear symmetry feature computed as proposed, has the property of orientation selectivity within each frequency channel. The algorithm is based on convolutions with simple separable filters and pixel-wise non-linear arithmetic operations. These properties allow highly parallel implementation, for example on a pyramid machine, yielding real time applications. Experiments based on test images of natural textures are presented.
Neural networks offer the potential for a quantum leap in the capabilities of imaging sensor systems. The critical neural network implementation factors are: weighted interconnect between all detector outputs;parallel...
详细信息
Neural networks offer the potential for a quantum leap in the capabilities of imaging sensor systems. The critical neural network implementation factors are: weighted interconnect between all detector outputs;parallel, linear processing of each detector output;fan-out to multiple (thousands) processing nodes per detector output;and the ability to independently change interconnect weights and processor node connections within the detector integration times. For a 128 × 128 pixel detector array, the number of desirable interconnects could be as high as 1015 per second, compared to the approximately 109 rates achieved presently with off-focal plane digital processors. Irvine Sensors Corporation (ISC) has conceived a new way of interconnecting 3-D focal plane readout modules and of laying out their component integrated circuits that appears to fulfill the very high interconnect rate requirements. This concept is described and interconnectivity and other performance attributes are discussed.
The Fourier transforming properties of coherent imagingsystems are well known and thoroughly developed. Optical spatial filtering techniques for coherent imagingsystems are also a well developed field of study. Most...
详细信息
The Fourier transforming properties of coherent imagingsystems are well known and thoroughly developed. Optical spatial filtering techniques for coherent imagingsystems are also a well developed field of study. Most of the advances in spatial filtering techniques were developed for systems with a known set of input characteristics. Once these filters were developed, they were static and unable to change with a new set of inputs exhibiting a different set of characteristics. This paper will present the theoretical basis and design of an adaptive coherent imaging system using a two-dimensional spatial light modulator (SLM) in the frequency plane. A computer controller is used as a feedback loop from the output plane back to the SLM. The filter implemented by the SLM is developed by simulated annealing. This iterative process enables the filter to adapt to a variety of inputs and automatically optimize given sufficient iteractions. This paper will also present feasible applications of this technique in imageprocessing and pattern recognition.
A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared ...
详细信息
ISBN:
(纸本)0819402931
A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared memory. The memory can be envisaged as an array of memory modules in the k-dimensional space, with each row of modules along a certain dimension connected to one bus. Each processor has access to one bus along each dimension. It is shown that these organizations are communication-efficient and can provide processor-time optimal solutions to a wide class of image and vision problems. In the two-dimensional case, the basic organization has n processors and an n × n memory array which can hold an n × n image, and it provides O(n) time solution to several image computations including: histograming, histogram equalization, computing connected components, convexity problems, and computing distances. Such problems also take O(n) time on a two-dimensional mesh with n2 processors. For the general k-dimensional case, a class of orthogonal data movement operations can be implemented on such organizations to yield processor-time optimal image and vision algorithms.
Space-based sensors currently transmit large amounts of raw data to their receiving stations, forcing large bandwidths to be used and large data archives maintained. One possible solution to this data management probl...
详细信息
Space-based sensors currently transmit large amounts of raw data to their receiving stations, forcing large bandwidths to be used and large data archives maintained. One possible solution to this data management problem is the use of more intelligent processing onboard the spacecraft. Neural networks are proposed as a trainable, mutable means of achieving sensor signal processing, sorting, and classification before infrared (IR) focal plane data leaves the cryogenic seal. Two questions arise: (1) can a neural network of appropriate size learn a sensor processing classification problem, and (2) will there be devices to implement that neural network as a smart focal plane. In this paper we concentrate on the first issue and show successful results in simulation for a massively parallel, distributed neural network solution to the closely spaced object (CSO) recognition problem, using a defocused IR sensor model with noise, and an uncooperative object or CSO at large distance. The back-propagation learning method is used to train the network. Architectural issues regarding the construction of a prototype neuromorphic focal plane device are also discussed.
A new generation of instrumentation will be developed for earth observation from a geosynchronous platform. The multisensor, multi-disciplinary requirements of this mission will require significant improvements in spa...
详细信息
A new generation of instrumentation will be developed for earth observation from a geosynchronous platform. The multisensor, multi-disciplinary requirements of this mission will require significant improvements in spatial and spectral resolution relative to instruments currently in use. Z-plane technology offers the potential for high resolution, visible and infrared imagingsystems with built-in calibration, data compression, image motion compensation and background suppression. Instrument needs are discussed, followed by examples of Z-plane technology solutions.
The proceedings contain 82 papers. The special focus in this conference is on Vector and parallelprocessing. The topics include: A decoupled data-driven architecture with vectors and macro actors;a novel paradigm of ...
ISBN:
(纸本)9783540530657
The proceedings contain 82 papers. The special focus in this conference is on Vector and parallelprocessing. The topics include: A decoupled data-driven architecture with vectors and macro actors;a novel paradigm of parallel computation and its use to implement simple high performance hardware;a bus-connected multiprocessor for a rete-based production system;a model for performance prediction of message passing multiprocessors achieving concurrency by domain decomposition;workloads, observables, benchmarks and instrumentation;a method for performance prediction of parallel programs;sparse matrixalgorithms for SUPRENUM;parallel givens factorization on a shared memory multiprocessor;study of a parallel lnference machine for paraliei execution of logic programs;parallel lmplementation of logic languages;prolog implementations on parallel computers;performance evaluation of parallel programs in parallel and distributed systems;the ELAN performance analysis environment;monitoring and debugging transputer-networks with NETMON-ii;an adaptive blocking strategy for matrix factorizations;factorizations of band matrices using level 3 BLAS;on the computation of breeding values;code parallelization for the LGDG large-grain dataflow computation;development of portable parallel programs with large-grain data flow 2;a latency tolerant code generation algorithm for a coarse grain dataflow machine;cedar fortran and its compiler;optimizing communication in superb;a design of performance-optimized control-based synchronization;interprocess analysis and optimization in the equational language compiler;transputer based distributed cartograptic lmage processing;parallel implementation of the convolution method in image reconstruction;analysis and design of circuit switching interconnection networks using 4x4 nodes.
暂无评论