检索结果-内蒙古大学图书馆

Computer Optics 2010年第3期34卷 360-366页

作者： Kulchin, Yuri Nikolaevich Zakasovskaya, Elena Vladimirovna Institute of Automation and Control Processes FEB RAS Russia Far Eastern National University Russia

The paper discusses tomography reconstruction of distributed physical fields by means of distributed fiber optical measuring systems (FOMN) for incomplete parallel schemes of measuring lines (ML) stacking. The approach is presented, consists in optimization of geometry of a measuring network for the purpose of the further application neural network methods of restoration of a full-image of investigated functions. Possibility of a choice and use of a suitable neural network from set of the several, in advance trained, neural networks of RBF-type is investigated. © 2010, Institution of Russian Academy of Sciences. All rights reserved.

关键词： Tomography

来源：评论

学校读者我要写书评

暂无评论

Scalability in Privacy-Preserving Data Aggregation for Wireless Sensor Networks

Scalability in Privacy-Preserving Data Aggregation for Wirel...

引用

International Symposium on parallel and distributed processing with Applications, ISPA

作者： Rabindra Bista Young-Sung Shin Jae-Woo Chang Department of Computer Engineering and Center of Advanced Image and Information Technology Chonbuk National University of Korea Cheongju Chungbuk South Korea

In this paper, we propose a new privacy preserving data aggregation scheme for WSNs. Our scheme applies additive property of complex numbers in order to combine sensor data and preserve data privacy during transmission to the sink node. In addition, for supporting scalability, we propose a novel mechanism in which a special set of real numbers are assigned to sensor nodes as their IDs so that a single bit is sufficient to hold ID of a sensor node during transmission of aggregated data to the sink node. For this, we, first, generate fixed size signatures for the IDs of all sensor nodes and then superimpose the signatures during data aggregation phase. By analytical evaluations, we show that our scheme is more scalable and energy efficient than the existing methods to achieve data privacy and transmit IDs of sensor nodes along with the aggregated data to the sink node.

关键词： Wireless sensor networks Data privacy Payloads Technical Activities Guide - TAG Encryption Tin Aggregates

来源：评论

学校读者我要写书评

暂无评论

On the parallelisation of MCMC by speculative chain execution

On the parallelisation of MCMC by speculative chain executio...

引用

IEEE International Symposium on parallel and distributed processing Workshops and Phd Forum (IPDPSW)

作者： Jonathan M. R. Byrd Stephen A. Jarvis Abhir H. Bhalerao Department of Computer Science University of Warwick Coventry UK

ISBN: (纸本)9781424465330

The increasing availability of multi-core and multiprocessor architectures provides new opportunities for improving the performance of many computer simulations. Markov Chain Monte Carlo (MCMC) simulations are widely used for approximate counting problems, Bayesian inference and as a means for estimating very high-dimensional integrals. As such MCMC has had a wide variety of applications in fields including computational biology and physics, financial econometrics, machine learning and image processing. One method for improving the performance of Markov Chain Monte Carlo simulations is to use SMP machines to perform 'speculative moves', reducing the runtime whilst producing statistically identical results to conventional sequential implementations. In this paper we examine the circumstances under which the original speculative moves method performs poorly, and consider how some of the situations can be addressed by refining the implementation. We extend the technique to perform Markov Chains speculatively, expanding the range of algorithms that maybe be accelerated by speculative execution to those with non-uniform move processing times. By simulating program runs we can predict the theoretical reduction in runtime that may be achieved by this technique. We compare how efficiently different architectures perform in using this method, and present experiments that demonstrate a runtime reduction of up to 35-42% where using conventional speculative moves would result in execution as slow, if not slower, than sequential processing.

关键词： Runtime Availability Computer architecture Computer simulation Monte Carlo methods Biological system modeling Computational modeling Bayesian methods Application software Computational biology

来源：评论

学校读者我要写书评

暂无评论

A distributed and Energy Efficient Algorithm for Data Collection in Sensor Networks

A Distributed and Energy Efficient Algorithm for Data Collec...

引用

International Conference on parallel processing Workshops (ICPPW)

作者： Sarah Sharafkandi David H. C. Du Alireza Razavi Department of Computer Science and Engineering University of Minnesota Twin city MN USA

In wireless sensor networks, collection of raw sensor data at a base station provides the flexibility to perform offline detailed analysis on the data which may not be possible with innetwork data aggregation. However, lossless data collection consumes considerable amount of energy for communication while sensors usually have limited energy. In this paper, we propose a distributed and Energy efficient algorithm for Collection of Raw data in sensor networks called DECOR. DECOR exploits spatial correlation to reduce the communication energy in sensor networks with highly correlated data. In our approach, at each neighborhood, one sensor shares its raw data as a reference with the rest of sensors without any suppression or compression. Other sensors use this reference data to compress their observations by representing them in the forms of mutual differences. In a highly correlated network, transmission of reference data consumes significantly more energy than transmission of compressed data. Thus, we first attempt to minimize the number of reference transmissions. Then, we try to minimize the size of mutual differences. We derive analytical lower bounds for both these phases and based on our theoretical results, we propose a two-step distributed data collection algorithm which reduces the communication energy significantly compared to existing methods. In addition, we modify our algorithm for lossy communication channels and we evaluate its performance through simulation.

关键词： Base stations distributed databases image color analysis Correlation Approximation methods Bridges Propagation losses

来源：评论

学校读者我要写书评

暂无评论

Star-ND: Multidimensional star-identification

Star-ND: Multidimensional star-identification

引用

AAS/AIAA Space Flight Mechanics Meeting

作者： Spratling iv, Ben Mortari, Daniele 746C H.R. Bright Building Aerospace Engineering Texas A and M University College Station TX 77843-3141 United States

ISBN: (纸本)9780877035602

One of the most computationally-complex steps in star-image-based autonomous spacecraft attitude-determination is star identification, in which the corresponding catalog-stars must be found for the directions to observed stars. Previous star identification methods have relied on 1) one-dimensional database searches, 2) tree-based multidimensional database searches, or 3) massively parallel processing hardware. The use of a multi-dimensional database search running in O(d + k)- time (where d is the number of dimensions and k is the number of correct answers), such as the K-Vector ND, promises to revolutionize the time required for star identification, if a set of star pattern parameters which are nearly uniformlydistributed, required by the K-Vector ND, are found. New star-pattern parameters, meeting the above specification are presented and found to be directly extensible to star patterns containing any number of stars greater than 2. The algorithm uses a star pattern composed of a selected star and its closest stars. The choice of this pattern causes a number of computational and reliability issues, and for that reason it is frequently considered to be the "wrong answer." methods of adding additional carefully-crafted "wrong" answers to the database to compensate for many of the issues are presented. Some issues require modification to the run-time algorithm, and the final run-time algorithm is presented. The finished Star-ND technique is evaluated both for execution-time performance in the presence of non-stars, and the growth of the database size for the false entries. Performance of the core algorithm is found to be as low as 1, 700 clock cycles. The required field-of-view size required for the algorithm to have enough stars to succeed in star identification is estimated and found to correspond with a simple mathematical expression.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

Low cost VLSI discrete wavelet transform and FIR filters architectures for very high-speed signal and image processing

Low cost VLSI discrete wavelet transform and FIR filters arc...

引用

IEEE International Conference on Cybernetic Intelligent Systems, CIS

作者： Mountassar Maamoun Rafik Bradai Abdelhamid Meraghni Rachid Beguenane Department of Electronics University of Blida Burgas Algeria LSIC Laboratory Ecole Normale Supérieure Kouba Algiers Algeria Department of Electrical and Computer Engineering Royal Military College of Canada Kingston Canada

This paper presents new VLSI architectures for finite impulse response (FIR) filters and discrete wavelet transform, intended for very high-speed signal and image processing. The proposed architectures, based on combining pipeline and parallel arithmetic methods, provide a new and very fast convolution approach with a reduced critical path. The key to this is a clever use of D-latches and multipliers which are efficiently distributed. Furthermore, an advanced discrete wavelet transform (DWT) approach, with an area-efficient architecture, is designed to produce one output in every clock cycle. As a result, a very high-speed is attained. The proposed structure can increase the work frequency (85%) at a low cost of additional hardware elements (55%). The systems are verified, using JPEG2000 coefficients filters, on Xilinx Field Programmable Gate Array (FPGA) devices.

关键词： Finite impulse response filter Computer architecture Adders Discrete wavelet transforms Low pass filters Convolution

来源：评论

学校读者我要写书评

暂无评论

3D Wave field phase retrieval from multi-plane observations

3D Wave field phase retrieval from multi-plane observations

引用

3DTV Conference

作者： Artem Migukin Vladimir Katkovnik Jaakko Astola Department of Signal Processing Tampere University of Technology Tampere Finland

We reconstruct a spatially distributed three-dimensional (3D) wave field from a number of intensity observations, obtained in different sensor planes, parallel to the object plane. The proposed algorithm can be treated as a multiple plane iterative Gerchberg-Saxton algorithm [1]. It is obtained from the best linear estimate of the complex-valued object distribution derived for the complex-valued observations. This estimator is modified for the intensity measurements in the sensor planes. The algorithm is studied by numerical experiments performed for amplitude and phase object distributions. It is shown that the proposed method allows reconstructing the whole 3D wave fields for different setup parameters. This technique can be applied for 3D imaging. The comparison versus the successive iterative method [2] shows an accuracy advantage of the proposed algorithm provided that the type of modulation in the object plane is known.

关键词： Signal processing algorithms image reconstruction Iterative algorithms Phase measurement Optical imaging Optical interferometry Iterative methods Holography Reconstruction algorithms Optical scattering

来源：评论

学校读者我要写书评

暂无评论

Pipelining parallel image compositing and delivery for efficient remote visualization

引用

JOURNAL OF parallel AND distributed COMPUTING 2009年第3期69卷 230-238页

作者： Wu, Qishi Gao, Jinzhu Chen, Zizhong Zhu, Mengxia Univ Memphis Dept Comp Sci Memphis TN 38152 USA Univ Pacific Dept Comp Sci Stockton CA 95211 USA Colorado Sch Mines Dept Math & Comp Sci Golden CO 80401 USA So Illinois Univ Dept Comp Sci Carbondale IL 62901 USA

Scientific datasets of large volumes generated by next-generation computational sciences need to be transferred and processed for remote visualization and distributed collaboration among a geographically dispersed team of scientists. parallel visualization using high-performance computing facilities is a typical approach to processing such increasingly large datasets. We propose an optimized image compositing scheme with linear pipeline and adaptive transport to support efficient image delivery to a remote client. The proposed scheme arranges an arbitrary number of parallel processors within a cluster in a linear order and divides the image into a carefully selected number of segments, which flow through the linear incluster pipeline and wide-area networks to the remote client consecutively. We analytically determine the segment size that minimizes the final image display time and derive the conditions where the proposed image compositing and delivery scheme outperforms the traditional schemes including the binary swap algorithm. In order to match the transport throughput for image delivery over wide-area networks to the pipelining rate for image compositing within the cluster, we design a class of transport protocols using stochastic approximation methods that are able to stabilize the data flow at a target rate. The experimental results from remote visualization of large-scale scientific datasets justify the correctness of our theoretical analysis and illustrate the superior performances of the proposed method. (C) 2008 Elsevier Inc. All rights reserved.

关键词： image compositing Pipelining Transport control Remote visualization parallel and distributed systems

来源：评论

学校读者我要写书评

暂无评论

Design of high-speed vision system and algorithms based on distributed parallel processing architecture for target tracking

Design of high-speed vision system and algorithms based on d...

引用

2009 7th Asian Control Conference, ASCC 2009

作者： Zhengtao, Zhang De, Xu Laboratory of Complex Systems and Intelligence Science Institute of Automation Chinese Academy of Sciences Beijing 100190 China

ISBN: (纸本)9788995605691

A high-speed stereovision system based on smart cameras is presented to track table tennis ball. A distributed parallel processing architecture is developed to improve the real-time performance of the system. A set of novel target recognition and tracking algorithm is also proposed to distinguish the ball from the changing background with less computation and good robustness. It takes about 6ms to capture an image and extract the ball. The trajectory of the ball is well measured in real-time. The experimental results are provided to verify the effectiveness of the proposed methods. ©2009 ACA.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

Non-rigid Registration for Large Sets of Microscopic images on Graphics Processors

引用

JOURNAL OF SIGNAL processing SYSTEMS FOR SIGNAL image AND VIDEO TECHNOLOGY 2009年第1-3期55卷 229-250页

作者： Ruiz, Antonio Ujaldon, Manuel Cooper, Lee Huang, Kun Ohio State Univ Biomed Informat Dept Columbus OH 43210 USA Univ Malaga Comp Architecture Dept E-29071 Malaga Spain

Microscopic imaging is an important tool for characterizing tissue morphology and pathology. 3D reconstruction and visualization of large sample tissue structure requires registration of large sets of high-resolution images. However, the scale of this problem presents a challenge for automatic registration methods. In this paper we present a novel method for efficient automatic registration using graphics processing units (GPUs) and parallel programming. Comparing a C++ CPU implementation with Compute Unified Device Architecture (CUDA) libraries and pthreads running on GPU we achieve a speed-up factor of up to 4.11x with a single GPU and 6.68x with a GPU pair. We present execution times for a benchmark composed of two sets of large-scale images: mouse placenta (16K x16K pixels) and breast cancer tumors (23K x62K pixels). It takes more than 12 hours for the genetic case in C++ to register a typical sample composed of 500 consecutive slides, which was reduced to less than 2 hours using two GPUs, in addition to a very promising scalability for extending those gains easily on a large number of GPUs in a distributed system.

关键词： Microscopic imaging image registration and segmentation Pattern analysis Feature detection Graphics processors High-performance computing

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：