A massively parallel resistive network for ''weak rod'': A double-layer architecture with nearest-neighbor connections

被引：15

作者：

Matsumoto, T

Kondo, K

机构：

[1] Department of Electrical Engineering, Waseda University, Tokyo 169, 3-4-1, Ohkubo, Shinjuku-ku

来源：

NEURAL NETWORKS | 1996年 / 9卷 / 03期

关键词：

weak rod; thin plate; line processes; vision; image processing; parallel networks; double-layer architecture; synaptic fuse;

D O I：

10.1016/0893-6080(95)00073-9

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A weak rod can be realized by a very natural double-layer network with feedback. While the second layer is linear, the first layer is significantly nonlinear. Elements associated with the first layer are called ''synaptic fuses''. An MRF (Markov random field) argument is used to eliminate the line variables. The network makes transparent those portions cut when a crease or a step edge is present. Copyright (C) 1996 Elsevier Science Ltd

引用

页码：523 / 541

页数：19

共 14 条

[1]

Blake A., 1987, Visual Reconstruction

[2]

HARRIS J, 1988, IEEE C NEUR INF PROC, V1, P687

[3]

HARRIS JG, 1991, THESIS CALTECH PASAD

[4]

HONGO S, 1992, T IEICE, V11, P1959

[5]

KAWATO M, 1991, 322 ATR

[6] IMAGE-PROCESSING REGULARIZATION FILTERS ON LAYERED ARCHITECTURE [J].

KOBAYASHI, H ;

MATSUMOTO, T ;

YAGI, T ;

SHIMMI, T .

NEURAL NETWORKS, 1993, 6 (03) :327-350

[7] AN ACTIVE RESISTOR NETWORK FOR GAUSSIAN FILTERING OF IMAGES [J].

KOBAYASHI, H ;

WHITE, JL ;

ABIDI, AA .

IEEE JOURNAL OF SOLID-STATE CIRCUITS, 1991, 26 (05) :738-748

[8] PARALLEL DISTRIBUTED NETWORKS FOR IMAGE SMOOTHING AND SEGMENTATION IN ANALOG VLSI [J].

LUMSDAINE, A ;

WYATT, J ;

ELFADEL, I .

PROCEEDINGS OF THE 28TH IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-3, 1989, :272-279

[9] REALIZATION OF THE WEAK ROD BY A DOUBLE-LAYER PARALLEL NETWORK [J].

MATSUMOTO, T ;

KONDO, K .

NEURAL COMPUTATION, 1994, 6 (05) :944-956

[10] SPATIAL VERSUS TEMPORAL STABILITY ISSUES IN IMAGE-PROCESSING NEUROCHIPS [J].

MATSUMOTO, T ;

KOBAYASHI, H ;

TOGAWA, Y .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (04) :540-569

← 1 2 →