NEW DIGITAL ARCHITECTURE OF CNN FOR PATTERN RECOGNITION

被引:14
作者
Raschman, Emil [1 ]
Zalusky, Roman [1 ]
Durackova, Daniela [1 ]
机构
[1] Slovak Tech Univ Bratislava, Dept Microelect, Bratislava 81219, Slovakia
来源
JOURNAL OF ELECTRICAL ENGINEERING-ELEKTROTECHNICKY CASOPIS | 2010年 / 61卷 / 04期
关键词
CNN; digital circuit; area consumption; pattern recognition; CELLULAR NEURAL-NETWORKS;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
080906 [电磁信息功能材料与结构]; 082806 [农业信息与电气工程];
摘要
The paper deals with the design of a new digital CNN (Cellular Neural Network) architecture for pattern recognition. The main parameters of the new design were the area consumption of the chip and the speed of calculation in one iteration. The CNN was designed as a digital synchronous circuit. The largest area of the chip belongs to the multiplication unit. In the new architecture we replaced the parallel multiplication unit by a simple AND gate performing serial multiplication. The natural property of this method of multiplication is rounding. We verified some basic properties of the proposed CNN such as edge detection, filling of the edges and noise removing. At the end we compared the designed network with other two CNNs. The new architecture allows to save till 86% gates in comparison with CNN with parallel multipliers.
引用
收藏
页码:222 / 228
页数:7
相关论文
共 13 条
[1]
BOUBAKER M, 2008, IJCSNS INT J COMPUTE, V8
[2]
CELLULAR NEURAL NETWORKS - APPLICATIONS [J].
CHUA, LO ;
YANG, L .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1988, 35 (10) :1273-1290
[3]
CELLULAR NEURAL NETWORKS - THEORY [J].
CHUA, LO ;
YANG, L .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1988, 35 (10) :1257-1272
[4]
Hnggi M., 2000, CELLULAR NEURAL NETW
[5]
JEON J, 2007, THESIS U RALEIGH N C
[6]
KVASNIAKA V, 1997, INTRO THEORY NEURAL
[7]
LARKIN D, 2006, LECT NOTES COMPUTER
[8]
LARSEN J, 1999, INTRO ARTIFICAL NEUR
[9]
Muthuramalingam A., 2008, INT J INFORM TECHNOL, V4
[10]
RASCHMAN E, 2008, P EL DEV SYST BRNO S, P322