检索结果-内蒙古大学图书馆

南京大学学报（自然科学版） 2016年第3期52卷 536-541页

作者：余双春袁杰南京大学电子科学与工程学院南京210093

提出将光声成像中的傅里叶域重建方法与分布式并行计算相结合的方法,应用于实际处理医学成像中,实现实时重建医学光声图像的目的.滤波反投影算法是光声图像重建的一个广泛使用的算法,但由于在时域重建的时间复杂度大,并不适合实时重建.... 详细信息

提出将光声成像中的傅里叶域重建方法与分布式并行计算相结合的方法,应用于实际处理医学成像中,实现实时重建医学光声图像的目的.滤波反投影算法是光声图像重建的一个广泛使用的算法,但由于在时域重建的时间复杂度大,并不适合实时重建.傅里叶域重建方法相对于滤波反投影算法具有时间复杂度上的优越性,且产生较少的伪影.并行处理技术,已经广泛应用于各类工程学科,其影响力及范围随着研究深入逐步扩大.分布式并行计算将提高图像处理速度,使实时重建医学图像成为可能.将MATLAB作为平台,实现了对医学图像在复频域的并行加速处理,给出了并行处理与串行处理的测试结果的比较.实验结果表明:傅里叶域重建方法与分布式并行计算的结合,使医学光声成像重建时间大大缩短.

关键词：光声成像傅里叶域重建并行处理技术分布式计算

来源：评论

学校读者我要写书评

暂无评论

Medical image segmentation on GPUs - A comprehensive review

引用

MEDICAL image ANALYSIS 2015年第1期20卷 1-18页

作者： Smistad, Erik Falch, Thomas L. Bozorgi, Mohammadmehdi Elster, Anne C. Lindseth, Frank Norwegian Univ Sci & Technol N-7491 Trondheim Norway SINTEF Med Technol N-7465 Trondheim Norway

Segmentation of anatomical structures, from modalities like computed tomography (CT), magnetic resonance imaging (MRI) and ultrasound, is a key enabling technology for medical applications such as diagnostics, planning and guidance. More efficient implementations are necessary, as most segmentation methods are computationally expensive, and the amount of medical imaging data is growing. The increased programmability of graphic processing units (GPUs) in recent years have enabled their use in several areas. GPUs can solve large data parallel problems at a higher speed than the traditional CPU, while being more affordable and energy efficient than distributed systems. Furthermore, using a GPU enables concurrent visualization and interactive segmentation, where the user can help the algorithm to achieve a satisfactory result. This review investigates the use of GPUs to accelerate medical image segmentation methods. A set of criteria for efficient use of GPUs are defined and each segmentation method is rated accordingly. In addition, references to relevant GPU implementations and insight into GPU optimization are provided and discussed. The review concludes that most segmentation methods may benefit from CPU processing due to the methods' data parallel structure and high thread count. However, factors such as synchronization, branch divergence and memory usage can limit the speedup. (C) 2014 The Authors. Published by Elsevier B.V.

关键词： Medical image Segmentation GPU parallel

来源：评论

学校读者我要写书评

暂无评论

parallel Graph Coloring for Manycore Architectures

Parallel Graph Coloring for Manycore Architectures

引用

International Symposium on parallel and distributed processing (IPDPS)

作者： Mehmet Deveci Erik G Boman Karen D. Devine Sivasankaran Rajamanickam Sandia National Laboratories Albuquerque NM

ISBN: (纸本)9781509021413

Graph algorithms are challenging to parallelize on manycore architectures due to complex data dependencies and irregular memory access. We consider the well studied problem of coloring the vertices of a graph. In many applications it is important to compute a coloring with few colors in near-lineartime. In parallel, the optimistic (speculative) coloring method by Gebremedhin and Manne is the preferred approach but it needs to be modified for manycore architectures. We discuss a range of implementation issues for this vertex-based optimistic approach. We also propose a novel edge-based optimistic approach that has more parallelism and is better suited to GPUs. We study the performance empirically on two architectures(Xeon Phi and GPU) and across many data sets (from finite element problems to social networks). Our implementation uses the Kokkos library, so it is portable across platforms. We show that on GPUs, we significantly reduce the number of colors (geometric mean 4X, but up to 48X) as compared to the widely used cuSPARSE library. In addition, our edge-based algorithm is 1.5 times faster on average than cuSPARSE, where it hasspeedups up to 139X on a circuit problem. We also show the effect of the coloring on a conjugate gradient solver using multi-colored Symmetric Gauss-Seidel method as preconditioner, the higher coloring quality found by the proposed methods reduces the overall solve time up to 33% compared to cuSPARSE.

关键词： Color Instruction sets Computer architecture Graphics processing units Algorithm design and analysis image color analysis Libraries

来源：评论

学校读者我要写书评

暂无评论

Connecting functional brain imaging and parallel distributed processing

引用

LANGUAGE COGNITION AND NEUROSCIENCE 2015年第4期30卷 380-394页

作者： Cox, Christopher R. Seidenberg, Mark S. Rogers, Timothy T. Univ Wisconsin Dept Psychol Madison WI 53706 USA

Functional neuroimaging and parallel distributed processing (PDP) theory, both introduced to cognitive science in the 1980s, led to influential research programmes that have proceeded in parallel with little mutual influence. The PDP approach advanced specific claims about the nature of neural representations that, perhaps surprisingly, have gone largely untested in functional brain imaging. One reason may be the widespread use of univariate statistical methods for analysing brain imaging data, which typically rely on assumptions that render them unable to detect distributed representations of the kind that PDP predicts. More recent multivariate methods for image analysis may be better suited to detecting such representations. In the current article, we consider why univariate methods have been insufficient to test PDP's representational claims, articulate some of the properties that neural representations ought to have if the PDP view is valid and then survey the recent neuroimaging literature for evidence that neural representations do or do not have these properties. The survey establishes that the PDP view of distributed representations has considerable evidential support. This analysis underscores the importance of understanding how the assumptions underlying methods for analysing functional imaging data constrain the kinds of questions that can be addressed. We then consider the implications for our developing understanding of the neural bases of cognition and for the design of future brain imaging studies.

关键词： fMRI distributed representations MVPA PDP

来源：评论

学校读者我要写书评

暂无评论

Maturity and disease detection in tomato using computer vision

Maturity and disease detection in tomato using computer visi...

引用

International Conference on parallel distributed and Grid Computing (PDGC)

作者： Tanvi Mehra Vinay Kumar Pragya Gupta Dept. of Electronics and Communication Engineering Thapar University Patiala India Dept. of Electronics and Communication Engineering Jaypee University of Information Technology Waknaghat Himachal Pradesh India

ISBN: (纸本)9781509036707

Our country has vast potential to come up as a cardinal exporter of agricultural produce, but lack of quick quality evaluation techniques, huge losses in processing and handling after harvesting, diseased crop etc. result in a lower contribution to global market. The tomato crop is often infected by a disease, where plant's leaves get covered with spots of colors dark brown with purple border and light grey center; termed as Septoria Leaf Spot. It causes the leaves to turn yellow, but most damage occurs due to loss of leaves by infection. In this paper, tomato maturity based on color and fungal infection in the tomato leaves is determined. Initially thresholding algorithm was performed to determine the maturity of tomato. To make the system more generalized and self-adapting a shift to k-means clustering algorithm is made. Finally a comparative analysis of both the methods was done to analyze which method is more suitable in different conditions. Also an unconventional machine vision system has been suggested that scrutinizes the leaves emerging out of the soil and depending upon leaf spots, it analyzes the nature of fungus and its depth into the stem of tomato. k-means algorithm along with thresholding is used for segmentation of image and eventually identifying fungus. The fungus part that is segmented, is then studied to derive the percentage of presence.

关键词： Decision support systems Clustering algorithms Diseases Agriculture image color analysis Algorithm design and analysis image segmentation

来源：评论

学校读者我要写书评

暂无评论

A Review on parallel Medical image processing on GPU 4

A Review on Parallel Medical Image Processing on GPU

引用

4th International Conference on Software Engineering and Computer Systems (ICSECS)

作者： Khor, Hui Liang Liew, Siau-Chuin Zain, Jasni Mohd. Univ Malaysia Pahang Fac Comp Syst & Software Engn Lebuhraya Tun Razak Kuantan 26300 Pahang Darul Ma Malaysia

ISBN: (纸本)9781467367226

An efficient implementation are necessary, as most medical imaging methods are computational expensive, and the amount of medical imaging data is growing. Graphic processing units (GPUs) can solve large data parallel problems at a higher speed than the traditional CPU, while being more affordable and energy efficient than distributed systems. This review investigates the use of GPUs to accelerate medical imaging methods. A set of criteria for efficient use of GPUs are defined. The review concludes that most medical image processing methods may benefit from GPU processing due to the methods' data parallel structure and high thread count. However, factors such as synchronization, branch divergence and memory usage can limit the speedup.

关键词： parallel computing Medical image processing GPU multicores

来源：评论

学校读者我要写书评

暂无评论

Local feature point extraction for quantum images

引用

QUANTUM INFORMATION processing 2015年第5期14卷 1573-1588页

作者： Zhang, Yi Lu, Kai Xu, Kai Gao, Yinghui Wilson, Richard Natl Univ Def Technol Coll Comp Sci & Technol Parallel & Distributed Proc Lab Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Coll Elect Sci & Engn Changsha 410073 Hunan Peoples R China Univ York Dept Comp Sci York YO10 5GH N Yorkshire England

Quantum image processing has been a hot issue in the last decade. However, the lack of the quantum feature extraction method leads to the limitation of quantum image understanding. In this paper, a quantum feature extraction framework is proposed based on the novel enhanced quantum representation of digital images. Based on the design of quantum image addition and subtraction operations and some quantum image transformations, the feature points could be extracted by comparing and thresholding the gradients of the pixels. Different methods of computing the pixel gradient and different thresholds can be realized under this quantum framework. The feature points extracted from quantum image can be used to construct quantum graph. Our work bridges the gap between quantum image processing and graph analysis based on quantum mechanics.

关键词： Quantum image Feature extraction Quantum adder Quantum image transformation

来源：评论

学校读者我要写书评

暂无评论

A novel three-dimension fast radar imaging method based on far-field-approximation

A novel three-dimension fast radar imaging method based on f...

引用

Progress in Electromagnetic Research Symposium (PIERS)

作者： Long Gang Wang Lianlin Li Tie Jun Cui Peking University Beijing Beijing CN Southeast University Nanjing Jiangsu CN

Three-dimension imaging and identification with radar has recently gained a lot of attention. Resulting from the over-simple imaging model and over-large computational complexity, the existing three-dimension radar imaging methods are limited in few scenarios. To break above bottlenecks, we introduce a novel physics-driven three-dimension fast radar imaging method based on far-field-approximation assumption, in which the whole imaging region would be decomposed with a series of overlapping-patches to accelerate the imaging speed in parallel. The proposed method has four key steps: e.g., first, the whole imaging region would be divided into a series of overlapping-patches, which indicates that the large-scale imaging problem is efficiently decomposed into a set of small-scale imaging sub-problems and can be solved in the parallel or distributed manner. Second, the dyadic Green's function based on far-field-approximation will be applied to construct system response function of radar imaging problem. Third, a dual transform is further introduced to turn the imaging problem into the problem of physical-driven image processing problem. In this way, the image processing system response functions for all overlapping-patches will be identical for fixed transmitters and receivers. In other words, the system response functions of overlapping-patches are independent of specific imaging regions. So few system response functions matrices need be memorized. At last, the first-order method is introduced to implement the imaging process of all overlapping-patches in parallel. The far-field-approximation imaging method drastically decreases the memory requirement while maintaining fast imaging speed and high imaging quality. Therefore, the proposed fast imaging method based on far-field-approximation is applicable to more large-scale imaging scenarios. Some selected simulation results are presented to demonstrate the state-of-the-art performance of the far-field-approximation method

关键词： Imaging Radar imaging image processing Electromagnetics Electromagnetic scattering Computer science Millimeter wave technology

来源：评论

学校读者我要写书评

暂无评论

基于远心照相的新型照相探头及用于多相搅拌槽测量

基于远心照相的新型照相探头及用于多相搅拌槽测量

引用

作者：王冠琦中国科学院大学(中国科学院过程工程研究所)

学位级别：硕士

多相反应器中分散相颗粒局部性质分布是反应器性能的精细反映,是多相反应器优化设计和工程放大的重要依据,也能够为多相反应器数学模型关键参数确定和数值模拟验证提供基础数据。目前多相反应器中分散相颗粒局部性质的实验测量方法都有... 详细信息

多相反应器中分散相颗粒局部性质分布是反应器性能的精细反映,是多相反应器优化设计和工程放大的重要依据,也能够为多相反应器数学模型关键参数确定和数值模拟验证提供基础数据。目前多相反应器中分散相颗粒局部性质的实验测量方法都有一定的局限性,尤其是对于气液固三相这种含有多种分散相颗粒的体系,还没有成熟的测量方法。本文开发了一种基于远心照相的新型照相探头,并发展了相应的测量方法,可用于两相及三相反应器中局部特性参数(如粒度和浓度等)的测量。测量时将照相探头安装在待测多相反应器中,按照预设的频闪光和CMOS(Complementary metal oxide semiconductor)同步控制模式即可拍照获得测量点处多相流动图像,采用image-Pro Plus图像处理软件结合相应的测量方法即可获得该点处分散相颗粒的粒度和浓度等性质信息。远心照相探头有四个优势:(1)远心镜头特有的平行光线使得拍摄的图像几乎不产生形变,与其他照相探头相比测量结果更准确;(2)探头尺寸小、远心镜头工作距离足够长,可用于测量大反应器内各个位置的局部特性参数;(3)可同时测量的局部特性参数包括颗粒(气泡、固体颗粒或液滴)的尺寸、相含率和界面接触面积;(4)克服了非侵入性照相法要求器壁透明等应用局限性问题。通过与游标卡尺测量的100个尺寸约1 mm的颗粒粒径结果对比表明,该实验照相探头的平均测量误差约是0.92%。在一种模拟的均匀膨胀床中测量颗粒浓度的结果表明,在不同膨胀高度条件下,该照相探头的测量误差最大是13.1%,平均测量误差为7.1%。这表明本文研制的远心照相探头在颗粒粒度和浓度测量中具有较高的准确性。利用远心照相探头拍摄了不同反光性的颗粒、具有不规则形貌的颗粒和含多个分散相的体系,表明在频闪光和CMOS同步控制模式下,或在特殊条件下辅以反光板(低亮度颗粒情况下),探头均能够清晰地捕捉到多相流动图像。在液-固搅拌槽内测量了局部相含率分布,测量结果与现有的认识定性相符;在气-液搅拌槽内测量了气泡尺寸分布,表明该探头能够测得100 μm以下的气泡,相对探针类方法具有明显优势;在气-液-固搅拌槽内同时测量了气泡和固体颗粒尺寸分布,表明该方法能应用于含有多个分散相体系的测量。

关键词：照相探头图像处理颗粒粒径局部相含率多相反应器

来源：评论

学校读者我要写书评

暂无评论

An algebraic parallel treecode in arbitrary dimensions 29

An algebraic parallel treecode in arbitrary dimensions

引用

29th IEEE International parallel and distributed processing Symposium (IPDPS)

作者： March, William B. Xiao, Bo Yu, Chenhan D. Biros, George Univ Texas Austin Inst Computat Engn & Sci Austin TX 78712 USA

ISBN: (纸本)9781479986484

We present a parallel treecode for fast kernel summation in high dimensions-a common problem in data analysis and computational statistics. Fast kernel summations can be viewed as approximation schemes for dense kernel matrices. Treecode algorithms (or simply treecodes) construct low-rank approximations of certain off-diagonal blocks of the kernel matrix. These blocks are identified with the help of spatial data structures, typically trees. There is extensive work on treecodes and their parallelization for kernel summations in three dimensions, but there is little work on high-dimensional problems. Recently, we introduced a novel treecode, ASKIT, which resolves most of the shortcomings of existing methods. We introduce novel parallel algorithms for ASKIT, derive complexity estimates, and demonstrate scalability on synthetic, scientific, and image datasets. In particular, we introduce a local essential tree construction that extends to arbitrary dimensions in a scalable manner. We introduce data transformations for memory locality and use GPU acceleration. We report results on the "Maverick" and "Stampede" systems at the Texas Advanced Computing Center. Our largest computations involve two billion points in 64 dimensions on 32,768 x86 cores and 8 million points in 784 dimensions on 16,384 x86 cores.

关键词： GPUs Kernel summation randomized linear algebra treecodes

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：