Image processing by neural networks

被引：12
作者：
Duranton, M
机构：
[1] Neural, Fuzzy
[2] Laboratoires d'Electron. Philips S., 94453 Limeil-Brévannes Cedex
来源：
IEEE MICRO | 1996年 / 16卷 / 05期
关键词：
D O I：
10.1109/40.540076
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
This fully programmable vector processor uses an array of 12 digital signal processors to perform up to 2 billion arithmetic operations per second and achieve a peak transfer throughput of 2 Gbytes/s for image-processing applications.
引用
收藏
页码：12 / 19
页数：8
相关论文
共 13 条
[1]  
DEJEAN C, 1994, P 1994 INT C SOL STA, P388
[2]  
Duranton M., 1996, Proceedings of the Fifth International Conference on Microelectronics for Neural Networks and Fuzzy Systems. MicroNeuro'96, P157, DOI 10.1109/MNNFS.1996.493786
[3]  
HAMMERSTROM D, 1991, VLSI ARTIFICIAL INTE
[4]  
*IBM SVI, 1993, ZER INSTR SET COMP Z
[5]  
KONDO Y, 1994, IEEE INT SOL STAT CI, P218
[6]   LNEURO 1.0 - A PIECE OF HARDWARE LEGO FOR BUILDING NEURAL NETWORK SYSTEMS [J].
MAUDUIT, N ;
DURANTON, M ;
GOBERT, J ;
SIRAT, JA .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (03) :414-422
[7]  
*NEST INC, 1994, NI1000 REC ACC US GU
[8]  
POUNTAIN D, 1995, BYTE FEB, P151
[9]  
RAMACHER U, 1991, PRELIMINARY SPECIFIC
[10]  
SACKINGER E, 1995, INT C ART NEUR NETW
12