检索结果-内蒙古大学图书馆

ARTIFICIAL NEURAL NETWORK AND image-processing USING THE ADAPTIVE SOLUTIONS ARCHITECTURE 2

CONF ON image processing ALGORITHMS AND TECHNIQUES

作者： BAKER, T Adaptive Solutions Inc. (United States)

ISBN: (纸本)0819405515

Adaptive Solutions' CNAPS architecture is a parallel array of digital processors. This design features a Single-Instruction Multiple-Data (SIMD) stream architecture. The architecture is designed to execute on- chip learning for Artificial Neural Network (ANN) algorithms with unprecedented performance. ANNs have shown impressive results for solving difficult image processing tasks. However, current hardware prevents many ANN solutions from being effective products. The CNAPS architecture will provide the computational power to allow real time ANN applications. Because of the high parallelism of the architecture,it is also ideal for digital image processing tasks. This architecture will allow high performance applications that combine conventional image processing methods and ANNs on the same system. This paper gives a brief introduction to the CNAPS architecture, and gives the system performance on implementation of neural network algorithms, and conventional image processing algorithms such as convolution, and 2D Fourier transforms.

关键词： image processing Clocks Evolutionary algorithms Convolution Artificial neural networks Fourier transforms Logic Neural networks Error analysis image filtering

来源：评论

学校读者我要写书评

暂无评论

NEW TRENDS IN MORPHOLOGICAL ALGORITHMS

NEW TRENDS IN MORPHOLOGICAL ALGORITHMS

引用

CONF ON NONLINEAR image processing 2

作者： VINCENT, L Harvard Univ. (United States)

ISBN: (纸本)0819405507

This paper describes some of the most recent algorithmic techniques of mathematical morphology. The classical parallel and sequential methods mostly involve numerous scannings of all the image pixels and are thus inefficient on conventional computers. To get rid of this drawback, the key idea is to consider, at each step, only the pixels whose value may be modified. Two classes of algorithms rely on this principle: the first one realizes an encoding of the object boundaries as loops which are then propagated in the image. The second one embraces the algorithms based on breadth- first image scanning enabled by queues of pixels. The algorithms belonging to these two families are extremely efficient and particularly suited to nonspecialized equipments, since they require random access to the pixels. Moreover, the 'customized' image scannings they are based on allow one to develop more accurate and flexible procedures: for example algorithms for computing exact Euclidean distance functions have been derived from the first class. On the other hand, the queue-based algorithms work in any kind of grid, in the Euclidean and geodesic cases, and extend to any dimension and even to graphs.

关键词： Algorithm development Binary data Nonlinear image processing Reconstruction algorithms Computing systems Computer programming image analysis Mathematical morphology Detection and tracking algorithms Video

来源：评论

学校读者我要写书评

暂无评论

image algebra preprocessor for the MasPar parallel computer

Image algebra preprocessor for the MasPar parallel computer

引用

image Algebra and Morphological image processing ii

作者： Meyer, Trevor E. Davidson, Jennifer L. Iowa State Univ. Ames IA USA

ISBN: (纸本)0819406961

The development of a preprocessor for image algebra on the MasPar computer, a SIMD processor array, is discussed. One of the parallel languages used on the MasPar is MPL, a parallel version of the programming language C. This machine and language were chosen because of the close correspondence between MPL and image algebra and, as a result, MPL is easily extended to include image algebra. The preprocessor consists of three primary components: the lexical analyzer, the parser, and the code generator. The lexical analyzer converts the input stream into tokens the parser recognizes. The parser checks the program for syntax *** code generator produces MPL source code which is then compiled and run on the MasPar. The architecture of the MasPar computer is reviewed, and in particular the structures used for routing data through the array are examined. The parallel language MPL is also reviewed, with attention given to the methods in which the extensions to C in MPL interact with the MasPar architecture. The close correspondence between MPL and image algebra is specifically discussed. The authors present the structure and development of the image algebra preprocessor, including possible future extensions. Examples of image algebra preprocessor code and the corresponding MPL code produced by the preprocessor are given.

关键词： Computer Systems, Digital

来源：评论

学校读者我要写书评

暂无评论

Examining the effects of CU/PE overlap and synchronization overhead when using the complete sums approach to image correlation

Examining the effects of CU/PE overlap and synchronization o...

引用

International Symposium on parallel and distributed processing (IPDPS)

作者： J.B. Armstrong M.A. Nichols H. Jay Siegel L.H. Jamieson Parallel Processing Laboratory School of Electrical Engineering Purdue University West Lafayette IN USA

A mixed-mode parallel machine's processing elements (PEs) are capable of operating in and switching between the SIMD and MIMD modes of parallelism. The paper analyzes various mappings of image correlation algorithms onto a mixed-mode parallel processing system. The trade-offs that exist between the SIMD and MIMD modes explain why some sequences of instructions are performed better in one mode than in the other and are the primary basis employed in comparing different mappings of a parallel algorithm onto a mixed-mode system.< >

关键词： parallel machines parallel processing image analysis parallel algorithms image processing Switches Laboratories Large-scale systems Pixel Smoothing methods

来源：评论

学校读者我要写书评

暂无评论

Approximation-based video tracking system

Approximation-based video tracking system

引用

image Algebra and Morphological image processing ii

作者： Deng, Keqiang Wilson, Joseph N. Univ. of Florida Gainesville FL USA

ISBN: (纸本)0819406961

This paper describes a two-dimensional target tracking system model and the image algebra specification of its algorithm. The visual tracking process in this model is divided into four subprocesses. Each target is abstracted as a parametrically described closed curve and the neighboring points of the curve. A description of the two-space motion of a target is obtained from the parameters defining the curve in sequential image frames, while curve construction and occlusion detection are accomplished using neighboring points of curve approximations from previous frames. The curve approximation is based on fitting the two-dimensional target boundary contour and is implemented as an error minimization problem. The simplest form of curve considered is an ellipse, but the target descriptions can be made more robust by adding more parameters as needed to describe more complex curves. The parameters of the curves identified in previous frames can be used as constraints in the error minimization process when occlusion is detected. The model can be applied to different types of targets by applying type-specific boundary estimation methods to each target's approximate boundary region. The image algebra description of this algorithm is concise and inherently parallel. This image algebra algorithm specification is of interest because it represents a complex vision activity rather than a simple filtering operation. In particular, the algorithm manipulates many small subimage regions specified by computing their point sets and gives a description of targets as a result, as is typical of vision algorithms. The implementation of the algorithm is described and its performance on a microcomputer vision system is characterized by giving two real-time tracking systems: a two-dimensional block tracking system and one that tracks human faces.

关键词： image processing

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional analysis framework and measurement methodology for imaging system noise

Three-dimensional analysis framework and measurement methodo...

引用

Infrared Imaging Systems: Design, Analysis, Modeling and Testing ii

作者： D'Agostino, John A. Webb, Curtis M. U.S. Army Cent. for Night Vision and Electro-Optics Alexandria VA USA

ISBN: (纸本)0819405973

Modern imaging sensors incorporate complex focal plane architectures and sophisticated post- detector processing. These advanced technical characteristics create the potential for multi- component noise generation which can exhibit effects temporally as well as along the vertical and horizontal image directions. Such complex three-dimensional (time, vertical, horizontal) noise cannot be adequately treated by previous mathematical analyses developed for simpler system designs where detector noise was predominant. In a parallel sense, earlier methods for noise measurement are no longer satisfactory. A new methodology has been developed at C2NVEO to characterize the noise patterns exhibited by advanced thermal imaging systems. The methods represents a significant expansion of the standard techniques to characterize thermal system noise. This paper explains the principles behind the 3-D noise methodology and the methods used. It also describes how this methodology is implemented in a laboratory measurement environment.

关键词： image Sensors

来源：评论

学校读者我要写书评

暂无评论

BDG-torus union graph-an efficient algorithmically specialized parallel interconnect

BDG-torus union graph-an efficient algorithmically specializ...

引用

International Symposium on parallel and distributed processing (IPDPS)

作者： N. Suri A. Mendelson D.K. Pradhan Department of Electrical and Computer Engineering University of Massachusetts Amherst MA Department of EE Technion-Israel Institute of Technology Israel

The binary de Bruijn graph (BDG) is a realizable alternative to the hypercube. An extension of the BDG by the edge set union with a torus is shown-union-graph (UG). This achieves graph capabilities/versatility comparable to the product shuffle (PS) graph and the hypercube within a fixed degree graph. The UG improves upon both in implementing pipelined and multi-phase algorithms. More importantly, the purpose being to design an algorithmically specialized interconnect, with the algorithmic features of a wide range of algorithms as well as direct architectural support for them, instead of simple providing for a set of graph embeddings in the interconnect. A set of examples demonstrate the UG's versatility in this aspect of algorithmic support.< >

关键词： Hypercubes Tree graphs Algorithm design and analysis Very large scale integration Costs Focusing Concurrent computing Finite element methods image analysis image processing

来源：评论

学校读者我要写书评

暂无评论

Effective load sharing in distributed real-time systems

Effective load sharing in distributed real-time systems

引用

International Symposium on parallel and distributed processing (IPDPS)

作者： K.-G. Shin C.-J. Hou Real-Time Computing Laboratory Dartment of Elec. Ennr. and ComDuter Science University of Michigan Ann Arbor MI USA

In a distributed real-time system, temporary uneven task arrivals among the nodes may cause some tasks to miss their deadlines even if the overall system has the capacity to meet the deadlines of all tasks. In this paper, an effective load sharing (LS) scheme is proposed as a solution to this problem. Upon arrival of a task at a node, the node determines whether or not it can complete the task in time under the minimum-laxity-first-served policy. If the task cannot be guaranteed or if guarantees of some other tasks are to be violated due to the insertion of this task into the existing schedule, the node looks up the list of loss-minimizing decisions, and determines the best node among a set of nodes in its physical proximity, called its buddy set, to which the task(s) may be transferred. This list of decisions is periodically updated using Bayesian decision analysis and prior/posterior state distributions. These probability distributions are derived from the information collected via time-stamped state-region change broadcasts within each buddy set. The performance of the proposed scheme is evaluated via simulation along with five other schemes. The proposed scheme is shown to outperform all but perfect LS scheme in (i) meeting task deadlines and (ii) tolerating the delays in state-information collection and task transfer.< >

关键词： Real time systems Broadcasting Delay Laboratories Bayesian methods Contracts

来源：评论

学校读者我要写书评

暂无评论

A Neural Controller for Mobile Robots

引用

IFAC Proceedings Volumes 1991年第9期24卷 271-275页

作者： M.B. Zaremba E. Porada Département d’informatique Université du Québec à Hull PQ Canada J8X 3X7

A neural controller for visually guided navigation of a mobile robot in the presence of dynamic obstacles and a vector field of attraction forces is presented. Discrete information, obtained from a raster-type sensor, is subjected to parallel distributed processing aiming at real-time detection of geometrical properties of visual representations of the obstacles. A characteristic feature of the neural architecture adopted here is processing of the input data by a hidden layer of dipole units. This novel approach allows the controller to perform the operations of edge detection and recognition of geometrical features within the controller’s structure rather than using visual data pre-processed by classical convolution-based methods. Dipole processing combined with the ability of our network to learn arbitrary associations very rapidly by self-programming permits the processing of large amounts of input data without compromising the efficiency of the learning phase. The capability of the network to extract and process some geometrical characteristics of the environment works for both local (limited by a field of view window) and global navigation. The behavior of the robot in the vicinity of obstacles can be tuned and modified by changing the number of recursively generated obstacle hulls, the size of the window, and the weights of the force components intervening in the control.

关键词： Neural nets robotics navigation control image processing intelligent machines modeling

来源：评论

学校读者我要写书评

暂无评论

HCL - A LANGUAGE FOR LOW-LEVEL image-ANALYSIS

引用

JOURNAL OF parallel AND distributed COMPUTING 1990年第3期8卷 231-244页

作者： PFEIFFER, JJ Department of Computer Science New Mexico State University Las Cruces New Mexico 88003-0001 USA

Pyramids have received a great deal of interest as a data structure for use in hierarchical computer vision. This paper describes a language for the description of pyramid algorithms, supporting conceptually parallel pattern-matching and arithmetic operations. The language is defined by adding constructs from a notation for pyramid algorithms called Hierarchical Cellular Logic to the general-purpose programming language C. Initially developed to support vision research on PCLIP ii, a cellular processor with a pyramidal interconnection topology, the language is also implementable on a wide range of parallel architectures.

关键词： image processing

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：