检索结果-内蒙古大学图书馆

Robustness of cellular neural networks in image deblurring and texture segmentation

INTERNATIONAL JOURNAL OF CIRCUIT THEORY AND APPLICATIONS 1996年第3期24卷 381-396页

作者： Sziranyi, T Analogical and Neural Computing Systems Research Laboratory Computer and Automation Institute Hungarian Academy of Sciences (MTA SZTAKI) pf. 63 H-1518 Budapest Hungary

Template parameters of cellular neural networks (CNNs) should be robust enough to random variability of VLSI tolerances and noise. Using the CNN for image processing, one of the main problems is the robustness of a given task in a real VLSI chip. It will be shown that very different tasks such as 2D or 3D deconvolution and texture segmentation can be solved in a real VLSI CNN environment without significant loss of efficiency and accuracy under low precision (about 6-8 bits) and random variability of the VLSI parameters. The CNN turns out to be very robust against template noise, image noise, imperfect estimation of templates and parameter accuracy. The parameters of a template are tuned using genetic learning. These optimized parameters depend on the precision of the architecture. It was found that about 6-8 bits of precision is enough for a complicated multilayer deconvolution, while only 4 bits of precision is enough for difficult texture segmentation in the presence of noise and parameter variances, The tolerance sensitivity of template parameters is considered for VLSI implementation. Theory and examples are demonstrated by many results using real-life microscopic images and natural textures.

关键词： Cellular neural networks

来源：评论

学校读者我要写书评

暂无评论

Feature extraction of postage stamps using an iterative approach of CNN

引用

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND computer SCIENCES 1996年第10期E79A卷 1741-1746页

作者： Kishida, J Rekeczky, C Nishio, Y Ushida, A Faculty of Engineering Tokushima University Tokushima-shi 770 Japan Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences 27 H-1111 Budapest Kende-u. 13-17 Hungary

In this article, a new analogic CNN algorithm to extract features of postage stamps in fray-scale images is introduced. The Gradient Controlled Diffusion method plays an important role in the approach. In our algorithm, it is used for smoothing and separating Arabic figures drawn with a color which is similar to the background color. We extract Arabic figures in postage stamps by combining Gradient Controlled Diffusion with nearest neighbor linear CNN template and logic operations. Applying the feature extraction algorithm to different test images it has been verified that it is also effective in complete segmentation problems.

关键词： cellular neural networks analogic algorithm image processing feature extraction gradient controlled diffusion

来源：评论

学校读者我要写书评

暂无评论

Real-life application case studies using CMOS 0.8 /spl mu/m CNN universal chip: analogic algorithm for motion detection and texture segmentation

Real-life application case studies using CMOS 0.8 /spl mu/m ...

引用

IEEE International Workshop on Cellular Nanoscale Networks and their Applications, (CNNA)

作者： P. Foldesy A. Zarandy P. Szolgay T. Sziranyi Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary

Using a 20/spl times/22 CNN Universal Machine chip two application case studies are presented. A new analogic CNN algorithm is shown to detect objects having larger size than a given value on black-and-white image sequences moving in a given range of direction and speed (17 /spl mu/s processing speed could be achieved). An extremely fast texture classification analogic algorithm is given next with approximately 2 /spl mu/s processing speed and with less than 5% misclassification error rate for 4 natural textures in a real-life testing environment.

关键词： computer aided software engineering Cellular neural networks Switches Logic Motion detection Turing machines Testing Prototypes Registers Thyristors

来源：评论

学校读者我要写书评

暂无评论

Picture segmentation with introducing an anisotropic preliminary step to an MRF model with cellular neural networks

Picture segmentation with introducing an anisotropic prelimi...

引用

13th International Conference on Pattern Recognition, ICPR 1996

作者： Szirányi, Tamás Czúni, László Department of Image Processing and Neurocomputing University of Veszprém 8200 Veszprém Egyetem út 10. Hungary Analogical and Neural Computing Laboratory Computer and Automation Institute Hungarian Academy of Sciences Kende u. 13-17 H- 1111 Budapest Hungary

ISBN: (纸本)081867282X

Due to the large computation power needed for Markovian random field (MRF) based image processing, new variations of the basic MRF model are implemented. The transportation of the model to the very fast cellular neural networks (CNN) gave new tasks and opportunities to improve the technique, since the CNN has a special local architecture. This CNN architecture can be implemented in real VLSI circuits of superior speed in image processing. A type of MRF image segmentation with modified metropolis dynamics (MMD) can be well implemented in the CNN architecture. In this paper we address the improvement of this existing CNN method by introducing anisotropic diffusion as the smoothing process in the model. We suggest that this new feature with the MRF representation will give a new approach to solving early vision problems in the future. © 1996 IEEE.

关键词： Cellular neural networks

来源：评论

学校读者我要写书评

暂无评论

Picture segmentation with introducing an anisotropic preliminary step to an MRF model with cellular neural networks

Picture segmentation with introducing an anisotropic prelimi...

引用

International Conference on Pattern Recognition

作者： T. Sziranyi L. Czuni Analogical and Neural Computing Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary Department of Image Processing and Neurocomputing University of Veszprém Hungary

关键词： Anisotropic magnetoresistance Cellular neural networks Image segmentation Image processing computer networks Pixel Labeling Smoothing methods Laboratories automation

来源：评论

学校读者我要写书评

暂无评论

CNN chip set architectures and the visual mouse

CNN chip set architectures and the visual mouse

引用

IEEE International Workshop on Cellular Nanoscale Networks and their Applications, (CNNA)

作者： T. Roska Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary Electronics Research Laboratory University of California Berkeley Berkeley CA USA

The cellular neural network (CNN) Universal Machine architecture when implemented in VLSI chips needs special interfaces to provide for efficient system performance both in time (speed) and space (image size). The CNN chip set architectures described solve this problem, including interfacing the possibly analog input sensors and digital output and control. Various forms of CNN Engines are presented embedding CNN chip sets. A new device, the Visual Mouse, a hand held visual supercomputer, is also presented which exploits the genuine features of CNN Engines.

关键词： Cellular neural networks Mice computer architecture Microprocessors Engines computer interfaces Spatiotemporal phenomena Buffer storage Laboratories Turing machines

来源：评论

学校读者我要写书评

暂无评论

On the detection of critical frequencies of a tool machine

On the detection of critical frequencies of a tool machine

引用

IEEE International Workshop on Cellular Nanoscale Networks and their Applications, (CNNA)

作者： P. Szolgay T. Szabo Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary Department of Mechanics University of Miskolc Egyetemváros Miskolc Hungary

An approximate solution is given for determining the ranges of allowed frequency of a mechanical model of a tool machine. The proposed method is based on the simulation of a measurement process of a real structure. The large computing power of cellular neural networks is used here to compute the transient response of a mechanical vibrating system.

关键词： Frequency Cellular neural networks computer networks Eigenvalues and eigenfunctions Computational modeling Finite element methods Mechanical systems Differential equations Laboratories automation

来源：评论

学校读者我要写书评

暂无评论

ACE: a digital floating point CNN emulator engine

ACE: a digital floating point CNN emulator engine

引用

IEEE International Workshop on Cellular Nanoscale Networks and their Applications, (CNNA)

作者： B. Feher P. Szolgay T. Roska A.G. Radvanyi T. Sziranyi M. Csapodi K. Laszlo L. Nemes I. Szatmari G. Toth P.L. Venetianer Measurement and Instrument Engineering Department Technical University of Budapest Budapest Hungary Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary

The architecture of ACE, a multiprocessor analogic cellular neural network (CNN) emulator engine consisting of 2 to 16 TMS320C40 floating point DSPs is introduced. The engine containing up to 512 Mbyte RAM (enough to store a 512/spl times/512/spl times/512 sized CNN cube) which can be controlled through its SCSI port. It can either accelerate the multilayer CNN simulator CNNM or be accessed directly from the high level, C-based analogic CNN language ACL to achieve the simulation speed of /spl sim/2.8 /spl mu/sec/cell/iteration/DSP for 3/spl times/3 linear templates.

关键词： Cellular neural networks Engines computer architecture Digital signal processing Very large scale integration Programmable logic arrays Computational modeling Logic arrays Signal processing algorithms Hardware

来源：评论

学校读者我要写书评

暂无评论

CNN model of the feature-linked synchronized activities in the visual thalamo-cortical system

CNN model of the feature-linked synchronized activities in t...

引用

IEEE International Workshop on Cellular Nanoscale Networks and their Applications, (CNNA)

作者： L. Orzo Z. Vidnyanszky J. Hamori T. Roska Lab. of Neurobiol. Semmuelweis Univ. Med. Sch. Budapest Hungary Laboratory of Neurobiology Semmelweis University Medical School Budapest Hungary Analogical and Neural Computing Systems Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary

In order to investigate whether thalamic synchronization may contribute to the binding of different cortical representations in visual information processing, we made a neuromorphic model of the thalamo-cortical loop, using the cellular neural network (CNN) model and simulator. It is demonstrated with the aid of this model that the cortico-thalamic feedback by synchronizing dLGN relay cell responses, provides a temporal code and this may induce correlated activities in the target cortical neuron population. Our results support the notion that using synchronization of temporally structured activities as general integration mechanism, the processing of visual information occurs simultaneously in the highly interconnected thalamo-cortical system.

关键词： Cellular neural networks Biological system modeling Neurons Information processing Relays Laboratories Neuromorphics Analytical models Anatomy computer networks

来源：评论

学校读者我要写书评

暂无评论

Diminishment and Enlargement sf Binary Bictures BTsing Slightlgr Space Variant Cellular neural Netwo

Diminishment and Enlargement sf Binary Bictures BTsing Sligh...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： C. Rekeczky A. Ushida T. Roska Department of Electrical and Electronks Engineering Faculty of Engineering University of Tokushima Tokushima Japan Analogical and Neural Computing Laboratory Computer and Automation Institute Hungarian Academy of Sciences (ATOMKI) Budapest Hungary

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：