Parameter adaptation in a simplified Pulse-Coupled Neural Network

被引：19

作者：

Székely, G ^{[1
]}

Lindblad, T ^{[1
]}

机构：

[1] MTA ATOMKI, H-4001 Debrecen, Hungary

来源：

NINTH WORKSHOP ON VIRTUAL INTELLIGENCE/DYNAMIC NEURAL NETWORKS: ACADEMIC/INDUSTRIAL/NASA/DEFENSE TECHNICAL INTERCHANGE AND TUTORIALS | 1999年 / 3728卷

关键词：

pulse-coupled neural network; PCNN; parameter adaptation; supervised learning;

D O I：

10.1117/12.343046

中图分类号：

V [航空、航天];

学科分类号：

08 ; 0825 ;

摘要：

In a general purpose pulse coupled neural network (PCNN) algorithm the following parameters are used: 2 weight matrices, 3 time constants, 3 normalization factors and 2 further parameters. In a given application, one has to determine the near optimal parameter set to achieve the desired goal. Here a simplified PCNN is described which contains a parameter fitting part, in the least squares sense. Given input and a desired output image, the program is able to determine the optimal value of a selected PCNN parameter. This method can be extended to more general PCNN algorithms, because partial derivatives are not required for the fitting. Only the slim of squares of the differences is used.

引用

页码：278 / 285

页数：8

共 7 条

[1]

[Anonymous], 1998, IMAGE PROCESSING USI

[2]

BROUSSARD RP, 1997, AFITDSGENG97J

[3] Feature Linking via Synchronization among Distributed Assemblies: Simulations of Results from Cat Visual Cortex [J].