检索结果-内蒙古大学图书馆

CONF ON stochastic and neural methods in signal processing, image processing, AND computer vision

作者： ZHUANG, XH WANG, T ZHANG, P Univ. of Missouri/Columbia (United States) Zhejiang Univ. (China) Univ. of Washington (United States)

ISBN: (纸本)081940697X

In Zhuang et al. (1988), a linear algorithm was presented to estimate a single instantaneous rigid motion from optic flow image point data. In order to obtain reasonable answers, however, the data must be quite accurate. This was shown by a lot of simulated experiments. As was well recognized, all machine vision feature extractors, recognizers, and matchers explicitly or implicitly needed for computing optic flow are unavoidably error prone and seem to make occasional errors which indeed are blunders. The realistic assumption for errors in optic flow should be a contaminated Gaussian noise which is a regular white Gaussian noise with probability 1 - (epsilon) plus an outlier process with probability (epsilon) , Huber (1981). Either the linear algorithm or the least-squares estimator are very sensitive to minor deviations from the Gaussian noise model assumption. In Haralick et al. (1989), the classical M-estimator was successfully applied to solve a single pose estimation from corresponding point data. However, lots of experiments conducted in Haralick et al. (1989) showed that the M-estimator only allowed a low proportion of outliers. For multiple pose segmentation and estimation, an estimator of high robustness is needed. A highly robust estimator called by the MF-estimator for general regression is presented and is applied to an important problem in computer vision, i.e., segmenting and estimating multiple instantaneous rigid motions from optic flow data. To be realistic, the observed or processed optic flow data are contaminated by various noises including outliers. Notationally, `MF' represents an abbreviation of `Model Fitting.' The MF- estimator is a result of partially modeling the unknown log likelihood function.

关键词： Machine vision computer vision technology Motion estimation image processing image segmentation Motion analysis stochastic processes signal processing Visual process modeling Motion models

来源：评论

学校读者我要写书评

暂无评论

FUZZY-LOGIC AND neural NETWORKS IN ARTIFICIAL-INTELLIGENCE AND PATTERN-RECOGNITION

FUZZY-LOGIC AND NEURAL NETWORKS IN ARTIFICIAL-INTELLIGENCE A...

引用

CONF ON stochastic and neural methods in signal processing, image processing, AND computer vision

作者： SANCHEZ, E Institut Mediterraneen de Technologie (France)

ISBN: (纸本)081940697X

With the use of fuzzy logic techniques, neural computing can be integrated in symbolic reasoning to solve complex real world problems. In fact, artificial neural networks, expert systems, and fuzzy logic systems, in the context of approximate reasoning, share common features and techniques. A model of Fuzzy Connectionist Expert System is introduced, in which an artificial neural network is designed to construct the knowledge base of an expert system from, training examples (this model can also be used for specifications of rules in fuzzy logic control). Two types of weights are associated with the synaptic connections in an AND-OR structure: primary linguistic weights, interpreted as labels of fuzzy sets, and secondary numerical weights. Cell activation is computed through min-max fuzzy equations of the weights. Learning consists in finding the (numerical) weights and the network topology. This feedforward network is described and first illustrated in a biomedical application (medical diagnosis assistance from inflammatory-syndromes/proteins profiles). Then, it is shown how this methodology can be utilized for handwritten pattern recognition (characters play the role of diagnoses): in a fuzzy neuron describing a number for example, the linguistic weights represent fuzzy sets on cross-detecting lines and the numerical weights reflect the importance (or weakness) of connections between cross-detecting lines and characters.

关键词： FUZZY LOGIC SYMBOLIC neural NETWORKS LINGUISTIC WEIGHTS WEIGHTED MINIMUM OF MAXIMA MEDICAL DIAGNOSIS HAND-WRITTEN CHARACTER RECOGNITION

来源：评论

学校读者我要写书评

暂无评论

SUPERRESOLUTION ALGORITHMS FOR A MODIFIED HOPFIELD neural NETWORK

引用

IEEE TRANSACTIONS ON signal processing 1991年第7期39卷 1516-1523页

作者： ABBISS, JB BRAMES, BJ FIDDY, MA UNIV LOWELL DEPT ELECT ENGNLOWELLMA 01854

The purpose of this paper is to describe the implementation of a superresolution (or spectral extrapolation) procedure on a neural network, based on the Hopfield model. This was first proposed by Abbiss et al. [1]. We show the computational advantages and disadvantages of such an approach for different coding schemes and for networks consisting of very simple two state elements as well as those made up of more complex nodes capable of representing a continuum. With the appropriate hardware, we show that there is a computational advantage in using the Hopfield architecture over some alternative methods for computing the same solution. We also discuss the relationship between a particular mode of operation of the neural network and the regularized Gerchberg-Papoulis algorithm.

关键词： Hopfield neural networks Biological neural networks Hardware computer architecture image restoration Cost function Extrapolation computer networks Biology computing Biological system modeling

来源：评论

学校读者我要写书评

暂无评论

A fast algorithm for a neocognitron neural network with back-propagation

A fast algorithm for a neocognitron neural network with back...

引用

stochastic and neural methods in signal processing, image processing, and computer vision 1991

作者： Qing, Pu Means, Robert W. HNC Inc. 5501 Oberlin Drive San DiegoCA92121 United States

The neocognitron is a neural network that consists of many layers of partially connected cells. In this paper, we propose a new neocognitron architecture called the Multi-layer NEOcognitron with BackPropagation learning (MNEOBP), and show that the original Neocongnitron that learned by backpropagation1'2 is a special case of what we propose here. The MNEOBP has a number of advantages: a) The algorithm for the MNEOBP is 4 times faster thatn the earlier algorithm since the number of cells calculated is reduced by a factor of 4. b) During the learning process, the mask (kernel) size is changed. This can speed up the training time by almost a factor of 3. c) The MNEOBP architecture can be implemented with a new digital neural network VLSI chip set called the vision Processor (ViP).3 The ViP exploits the convolution structure of the network and can processes a single 32 x 32 input layer in only 25.6 microseconds with an 8 x 8 receptive field kernel. © 1991 SPIE. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

25th Asilomar Conference on signals, Systems & computers

引用

25th Asilomar Conference on signals, Systems & computers Part 1 (of 2)

ISBN: (纸本)0818624701

The following topics are dealt with: signal processing and coding for digital storage systems;implementation of algorithms and computations;transforms and convolution;radar and sonar signal processing;knowledge-based signal processing;adaptive filtering;matrix-based signal processing;speech processing;spectral estimation;arithmetic for digital signal processing (DSP);DSP implementations;image coding;neural networks for signal processing;VLSI implementations;evolutionary programming;array signal processing;topics in signal processing;blind adaptive processing;quadrature-mirror filters;wavelets and applications;biologically motivated signal processing;military applications;communication and estimation;signal subspace methods;adaptive beamforming;digital filters;image processing;detection and characterization of transient signals;neural network applications;communication systems;computer vision;and testing and fault tolerance.

关键词： signal processing

来源：评论

学校读者我要写书评

暂无评论

24th Asilomar Conference on signals, Systems and computers

24th Asilomar Conference on Signals, Systems and Computers

引用

24th Asilomar Conference on signals, Systems and computers Part 2 (of 2)

The volume contains 107 Conference papers. The main topics covered include image processing, electrooptical sensor signal processing, nonlinear systems, knowledge-based signal processing, speech applications, signal processing and coding for digital storage systems, multiprocessor architectures, adaptive filters, array signal processing, computer vision, residue number systems, fault tolerance and testing, signal processing for digital facsimile, VLSI issues in algorithms and computations, signal subspace methods, neural nets in signal processing, evolutionary programming, DSP implementations, and applications of time-frequency signal representations.

关键词： signal processing

来源：评论

学校读者我要写书评

暂无评论

Visual Communications and image processing '91: image processing

Visual Communications and Image Processing '91: Image Proces...

引用

Visual Communications and image processing '91: image processing Part 2 (of 2)

ISBN: (纸本)0819407437

Proceedings incorporates 45 papers that are subdivided into seven sessions dealing with: image segmentation and classification;digital image processing in medicine;imqage sequence restoration and filtering;digital image processing algorithms;applications of digital image processing;and neural networks in image processing. Topics discussed include: object identification, shape recognition, associative neural networks, wavelet-type functions, color images, space-filling molecular models, image generation, teleconferencing systems, video browsing, chip resistors, quality inspection, picture processing and painting, computer vision;image retrieval, digital halftoning, ASIC architecture, video signal processing, Gabor decomposition, VLSI realization, Gaussian operators, Fourier and Gabor transforms, sine and cosine transforms, programmable processors, multidimensional signal processing, IDS filters, affine groups, Rader's algorithm, autocorrelation, signal reconstruction, heart movies, motion-compensated filtering, video sequences, median filter, adaptive methods, cross-entropy algorithm, Prosthetic sockets, cell shape analysis, CT images, venous beading, texture recognition, Chinese-English documents, and block segmentation.

关键词： image processing

来源：评论

学校读者我要写书评

暂无评论

Nonlinear analog networks for image smoothing and segmentation

引用

Journal of VLSI signal processing 1991年第1-2期3卷 53-68页

作者： Lumsdaine, A. Wyatt Jr., J.L. Elfadel, I.M. Research Laboratory of Electronics Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge 02139 MA United States

image smoothing and segmentation algorithms are frequently formulated as optimization problems. Linear and nonlinear (reciprocal)resistive networks have solutions characterized by an extremum principle. Thus, appropriately designed networks can automatically solve certain smoothing and segmentation problems in robot vision. This paper considers switched linear resistive networks and nonlinear resistive networks for such tasks. Following [1] the latter network type is derived from the former via an intermediate stochastic formulation, and a new result relating the solution sets of the two is given for the "zero temperature" limit. We then present simulation studies of several continuation methods that can be gracefully implemented in analog VLSI and that seem to give "good" results for these nonconvex optimization problems. © 1991 Kluwer Academic Publishers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proceedings, Advanced computer Technology, Reliable Systems and Applications

Proceedings, Advanced Computer Technology, Reliable Systems ...

引用

Proceedings of the 5th European computer Conference on Advanced computer Technology, Reliable Systems and Applications - CompEuro 91

ISBN: (纸本)0818621419

The following topics are dealt with: computer architecture;dedicated image processing systems;packet switched network computer communications;systems and software;models and algorithms;computer communications;software development support;VLSI design, production, and testing TPG (test pattern generation) and fault simulation;protocols;access methods for multimedia DBMS (database management systems);Petri nets;digital signal processing;database specification and methods;fault tolerance techniques;neural networks;complex VLSI chips;processor arrays;database languages and models;CAD and applications. Abstracts of individual papers can be found under the relevant classification codes in this or other issues.

关键词： computers

来源：评论

学校读者我要写书评

暂无评论

neural NETWORKS IN BANDWIDTH COMPRESSION 14

NEURAL NETWORKS IN BANDWIDTH COMPRESSION

引用

CONF ON APPLICATIONS OF DIGITAL image processing 14

作者： HABIBI, A The Aerospace Corp. (United States)

ISBN: (纸本)0819406953

The first objective of this study is to analyze the use of neural networks for bandwidth compression. This is achieved by developing a neural network algorithm using the Kohonen self-organization technique to perform vector quantization. The second objective of the study is to combine the neural network vector quantizer with a DPCM encoder for more efficient bandwidth compression. The bandwidth compression techniques are simulated and their performance is evaluated using one-dimensional wideband signals. Vector quantization (VQ) has proved to be an efficient method for bandwidth compression of both one-dimensional signals and imagery data. The reason VQ is not utilized in many practical applications is due to the fact that the performance of VQ becomes superior to other techniques such as transform coding and DPCM only for large vector dimensions. Large vector dimensions also increase the number of computations (per sample) and the memory requirements of the VQ; the increase is an exponential function of the vector dimension. For this reason, in this study VQ has been used in combination with other methods of bandwidth compression so that using a small vector dimension still improves the overall system performance. neural networks present a parallel approach to data classification that may simplify the architecture of the classifier in the VQ, thus making vector quantizers with large dimensions more practical. In this study, we have developed a neural network algorithm for vector quantization which is based on Kohonen self-organization technique. In the following we discuss the neural network classifiers and their utilization in vector quantization of the DPCM encoders. Simulation results showing the system performance of these systems for one-dimensional modulated signals is presented and the results are discussed.

关键词： neural networks Quantization Algorithm development computer programming Evolutionary algorithms computer simulations Distortion signal to noise ratio Aerospace engineering Modulation

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：