检索结果-内蒙古大学图书馆

International Conference on pattern recognition

作者： H. Bischof A. Pinz W.G. Kropatsch Institute for Automation Department for Pattern Recognition and Image Processing Technical University of of Vienna Vienna Austria

The interpretation of neural network behavior is of particular interest in neural network research. Visualization methods provide the necessary means to simultaneously analyze the huge amount of information hidden in the network. The authors propose a framework for visualization methods suited for feed forward neural networks. The basic idea is to use the spatial information available outside the network to arrange the data to be visualized (weights, activations of units) in the spatial domain of the display. Several examples which illustrate the proposed framework are presented.< >

关键词： Neural networks Data visualization pattern recognition Information analysis Feedforward systems Automation image processing Computer displays Information processing Computer vision

来源：评论

学校读者我要写书评

暂无评论

Combining pyramidal and fractal image coding

Combining pyramidal and fractal image coding

引用

International Conference on pattern recognition

作者： W.G. Kropatsch M.A. Neuhauser I.J. Leitgeb H. Bischof Department for Pattern Recognition and Image Processing Institute for Automation Technical University of of Vienna Vienna Austria

Compared with the richness in detail of images generated by iterated function systems, the memory requirement of them is very low. Displaying these images is very time consuming, and currently there is no method published to derive an iterated function system directly from a given image. This is known as the inverse problem. The authors introduce discrete transformations and direct computation of a discrete attractor with deterministic noniterative algorithms. This results in an essential saving of time. Furthermore they present a solution of the inverse problem in the one dimensional discrete space.< >

关键词： Fractals image coding Inverse problems Extraterrestrial measurements pattern recognition image processing Automation image generation Neural networks image recognition

来源：评论

学校读者我要写书评

暂无评论

Information fusion in image understanding

Information fusion in image understanding

引用

International Conference on pattern recognition

作者： A. Pinz R. Bartl Department for Pattern Recognition and Image Processing Institute for Automation Wien Austria Institute for Surveying and Remote Sensing University of Bodenkultur Wien Austria

Computer vision and image understanding processes are not very robust; small changes in exposure parameters or in internal parameters of algorithms can lead to significantly different results. A combination (fusion) of these results is profitable. The authors introduce an extended fusion concept dealing with different sources of information at external (world, scene, image) and internal (image description, scene description) levels and define the process of fusion. Each level requires its own procedure of quality measure and information fusion in order to yield a combination of components from several sources. Related work in the field is reviewed. Examples from the authors' own work cover remote sensing (improvement of classification results by fusion at the image level), medical image processing of ocular fundus images (automatic control point selection by fusion at the image description level) and the interpretation of Billard scenes (object identification by fusion at the scene description level).< >

关键词： Layout Computer vision Remote sensing image processing Robustness Application software pattern recognition Automation Lead Information resources

来源：评论

学校读者我要写书评

暂无评论

SAM investigations of the segregation of sulphur in polycrystalline α-iron

SAM investigations of the segregation of sulphur in polycrys...

引用

作者： Eisl, M.M. Reichl, B.M. Böhmig, S.D. Störi, H. Institut für Allgemeine Physik University of Technology Vienna Wiedner Hauptstraße 8-10 ViennaA-1040 Austria Institute for Automation Dept. for Pattern Recognition and Image Processing University of Technology Vienna Wiedner Hauptstr. 8-10 ViennaA-1040 Austria

Scanning Auger microscopy (SAM) and AES investigations of the segregation of non-metal impurity atoms (sulphur, nitrogen, carbon and phosphorus) in polycrystalline high-purity (>99.99%) α-iron samples were performed in the temperature range 500-800°C. At the beginning of the heating, nitrogen segregates homogeneously and quickly but in the equilibrium state the whole surface is covered with nearly half a monolayer of sulphur. During the first measurements, the segregation of sulphur was ruled by a combined bulk and grain boundary diffusion mechanism with a t1/2 law, resulting in a net-shaped structure of the surface concentration of sulphur. After repeated heating and sputtering, the contribution of grain boundary diffusion becomes smaller and the segregation rate of sulphur decreases as a consequence of sulphur depletion of the bulk. The segregation kinetics of the individual grains differ strongly owing to the differences between the extent of sulphur depletion of the grains and can be described very well by a t3/2 law in the initial stage of the segregation process. In a later phase, surface diffusion of sulphur from the grains, which have almost reached the equilibrium sulphur coverage, leads to an almost homogeneous sulphur coverage of the surface with only small and diffuse lateral variations. © 1992 John Wiley & Sons Ltd.

关键词： Sulfur

来源：评论

学校读者我要写书评

暂无评论

Tools for dynamic network structures: GRAPE grammars

Tools for dynamic network structures: GRAPE grammars

引用

International Joint Conference on Neural Networks (IJCNN)

作者： R. Freund B. Haberstroh H. Bischof Institute for Computer Languages Technical University of of Vienna Vienna Austria Institute for Information Systems Technical University of of Vienna Vienna Austria Institute for Pattern Recognition and Image Processing Technical University of of Vienna Vienna Austria

A theoretical framework based on attributed elementary programmed graph grammars (GRAPE grammars) that turns out to be equally well-suited for describing network dynamics, weight learning algorithms, and topology learning algorithms is proposed. It is shown how GRAPE grammars can be used to model neural networks. Special emphasis is placed on topology learning. It is concluded that GRAPE grammars offer a great potential for neural networks, giving the possibility of a common language suited for all kinds of neural networks.< >

关键词： Pipelines Network topology Neural networks Production Heuristic algorithms Dynamic programming Computer languages Electronic mail pattern recognition Formal languages

来源：评论

学校读者我要写书评

暂无评论

A THREE-LAYER ADAPTIVE NETWORK FOR pattern DENSITY ESTIMATION AND CLASSIFICATION

引用

International Journal of Neural Systems 1991年第3期2卷 211-220页

作者： Jian-xiong Wu Chorkin Chan Department of Computer Science University of Hong Kong Hong Kong Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University P.R.C

Based on the assumption that most probability densities in real life can be approximated by a mixture of Gaussian densities, we propose here a three-layer adaptive network with each neuron in the lower hidden layer representing a Gaussian basis function (covariance matrix equal to where I is a unit matrix) to estimate various probability densities and serve as a Bayes classifier. The width of the basis function may be the same for all neurons in this layer or it may vary from one neuron to another. This paper investigates the effectiveness of the network for both cases and presents a localized learning algorithm to adjust the network parameters. The network was trained with artificial data derived from known mixtures of memoryless Gaussian sources as well as exponential and Gamma densities. The performance of the network as a pattern density estimator was measured in terms of the relative difference between the target probability density function (p.d.f.) which generates the training and testing data and the network output representing the estimation. Samples from two mixtures corresponding to two classes were used to test the network capability as a classifier by comparing its error rate against that of a Bayes classifier. Both one- and two-dimensional cases were explored. The successfulness of the network depended on how well the target p.d.f.’s were represented by the training samples, the number of hidden neurons employed in the network and how thoroughly the network was trained. It was also found that allowing each basis function to have an independent width had a predominant effect on the network performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A new adaptive classified transform coding method (image coding)

A new adaptive classified transform coding method (image cod...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Wen-Jun Zhang Song-Yu Yu Hong-Bin Chen Institute of Image Processing & Pattern Recognition Shanghai Jiaotong University Shanghai China

An efficient discrete cosine transform technique using a new adaptive feature for bandwidth compression is described. Taking account of human visual characteristics and transform coefficient statistics, the higher activity region is further classified into four subclasses according to four proposed basic patterns, while the lower activity region is assigned to four subclasses according to its AC-energy (image activity) distribution. Computer simulations shows that the proposed adaptive coder exhibits a performance improvement of 1 dB or more over conventional adaptive coders at the same coding rates.< >

关键词： Transform coding image coding Humans Discrete transforms image reconstruction Degradation Bandwidth Bit rate image storage Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

An improved method for shape from shading

An improved method for shape from shading

引用

International Conference on pattern recognition

作者： Wenjun Zhang Yuanhua Zhou Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai China

ISBN: (纸本)0818608781

An improved method for shape from shading is presented. With introduction of the adaptive attenuated factors, the results of the initial iterations fall into the region of the solution values as much as possible. Simulated tests show that the method makes a notable improvement over the well-known approach proposed by K. Ikeuchi and B.K.P. Horn (1981), not only on the correctness of the solution but also on the speed of the convergence. The case of the actual object is also discussed. The result is satisfactory.< >

关键词： Shape image sampling Equations image processing pattern recognition Testing Computer vision Stereo vision Mathematical model image generation

来源：评论

学校读者我要写书评

暂无评论

The Del /sup 2/G performance in the presence of noise

The Del /sup 2/G performance in the presence of noise

引用

International Conference on pattern recognition

作者： Xie Zhaohui Wu Lide Institute of Image Processing & Pattern Recognition Shanghai Jiaotong University Shanghai China

ISBN: (纸本)0818608781

The Laplacian of Gaussian operator, Del /sup 2/G, is very important as an edge detector in the theory of computer vision. The bias of zero-crossing and output signal-to-noise-ratio of Del /sup 2/G under the models of four typical kinds of edges corrupted by white noise are given, and these theoretical results are confirmed by experiments. The relations among bias of zero-crossing, output and input signal-to-noise-ratio and parameter sigma of Del /sup 2/G are presented.< >

关键词： Detectors Filtering White noise Proposals Performance analysis image edge detection Pulse shaping methods Kernel Shape

来源：评论

学校读者我要写书评

暂无评论

A knowledge-based vision system for identification overlapping objects

A knowledge-based vision system for identification overlappi...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Yang Ming-chien Shi Peng-fei Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai China

ISBN: (纸本)7800030393

The intention of this paper is to help bridging the gap between knowledge base and computer vision system. A knowledge-based vision system for identification of overlapping objects is presented. The authors place emphasis on the reasoning strategy based on knowledge base for recognizing of occluded workpieces to provide information with an education Robot. The experimental results are given and some problem are discussed.

关键词： Machine vision Shape Noise shaping Layout Object recognition Character recognition Solid modeling pattern recognition Face recognition

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：