SHAPE APPROXIMATION OF ARC PATTERNS USING DYNAMIC NEURAL NETWORKS

被引：3

作者：

PARUI, SK

DATTA, A

PAL, T

机构：

[1] Electronics and Communication Sciences Unit, Indian Statistical Institute, Calcutta, 700 035

[2] Computer and Statistical Service Centre, Indian Statistical Institute, Calcutta, 700 035

来源：

SIGNAL PROCESSING | 1995年 / 42卷 / 02期

关键词：

ARC PATTERNS; SELF ORGANIZATION; DYNAMIC NEURAL NET; SKELETONIZATION;

D O I：

10.1016/0165-1684(94)00130-R

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

A shape representation technique for two-dimensional patterns using a dynamic variation of Kohonen's self-organizing feature maps is discussed. In Kohonen's map, the number of processors is fixed and is to be specified a priori. This number, if not properly chosen, can cause either wastage or shortage of processors. The problem is overcome, in this paper, by incorporating dynamic growing and shrinking capabilities in the network. Shape of only are patterns is considered here.

引用

页码：221 / 225

页数：5

共 7 条

[1]

Carpenter, Grossberg, Neural Networks for Vision and Image Processing, (1992)

[2]

Eckmiller, Malsburg, Neural Computers, (1989)

[3]

Fritzke, Let it grow — Self-organizing feature maps with problem dependent cell structure, Artificial Neural Networks, 1, pp. 403-408, (1991)

[4]

Kohonen, Self-Organization and Associative Memory, (1989)

[5]

Lee, Patterson, A hybrid neural network vision system for object identification, Artificial Neural Networks, 1, pp. 69-74, (1991)

[6]

Pao, Adaptive Pattern Recognition and Neural Networks, (1989)

[7]

Sabourin, Mitiche, Modeling and classification of shape using a Kohonen associative memory with selective multiresolution, Neural Networks, 6, pp. 275-283, (1993)

← 1 →