𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Theory, design and applications of Cellular Neural Networks

✍ Scribed by Roska, Tamas; Rodriguez-Vazquez, Angel; Vandewalle, Joos


Publisher
John Wiley and Sons
Year
1998
Tongue
English
Weight
35 KB
Volume
26
Category
Article
ISSN
0098-9886

No coin nor oath required. For personal study only.

✦ Synopsis


This special issue on CNN addresses three areas: theory, design, and applications. It is clear from the table of contents that most of the advances reported here are, explicitly or tacitly, influenced by the existence of the programmable CNN universal machine chips. A new programming discipline, the analogic CNN algorithmic thinking, is emerging where the elementary instruction is a non-linear spatiotemporal operator defined by a CNN cloning template. Circuit theory, systems aspects, and algorithm design are developing hand-inhand.

In the Theory section, the realizability of large neighbourhood templates on 3;3 neighbourhood machines, as well as a unified characterization of multilayer higher-order CNN are presented. In addition to these, the synthesis of three types of operators are presented: (i) arbitrary local Boolean functions, (ii) invertible Boolean operators, and (iii) patterns and autowaves.

In the section on Design the formulation of the 'art' of template design is presented for propagating binary operators, including the famous connectivity problem, an analogic CNN algorithmic frame is presented for stereo problems, and a CAD environment is described for the VLSI CNN chip designer. Robustness is a key issue when implementing algorithms on CNN Universal programmable chips. The two papers in this area cover the theoretical as well as the practical aspects.

The main theme of the Applications section is video coding and analysis of noisy motion pictures. In addition, a layout error detection algorithm is presented.

Physical implementation, including computing infrastructure, and the biologically inspired aspects are not covered in this special issue. These areas are, however, the main themes of a companion special issue in the IEEE Transactions on Circuits and Systems.


πŸ“œ SIMILAR VOLUMES


A rigorous design method for binary cell
✍ Fajfar, Iztok; Bratkovič, Franc; Tuma, Tadej; Puhan, Janez πŸ“‚ Article πŸ“… 1998 πŸ› John Wiley and Sons 🌐 English βš– 85 KB πŸ‘ 1 views

In order to be able to take full advantage of the great application potential that lies in cellular neural networks (CNNs) we need to have successful design and learning techniques as well. In almost any analogic CNN algorithm that performs an image processing task, binary CNNs play an important rol

Heteroassociative memories via cellular
✍ Brucoli, Michele; Carnimeo, Leonarda; Grassi, Giuseppe πŸ“‚ Article πŸ“… 1998 πŸ› John Wiley and Sons 🌐 English βš– 124 KB πŸ‘ 1 views

In this paper a synthesis procedure for heteroassociative memories using Cellular Neural Networks (CNNs) is presented. The suggested method, by assuring the condition of symmetry of the interconnection matrix, guarantees the complete stability of the designed network, besides providing that all the

On the separating capability of cellular
✍ Osuna, J. A.; Moschytz, G. S. πŸ“‚ Article πŸ“… 1996 πŸ› John Wiley and Sons 🌐 English βš– 456 KB πŸ‘ 2 views

The cellular neural network is able to perform different image-processing tasks depending on the template values, i.e. the network parameters, used. In the case of linear templates the parameter space is divided into different regions by hyperplanes. Every region is associated with a task, such that