General purpose neuro-image processor architecture

Gunhee Han, Edgar Sanchez-Sinencio

Research output: Contribution to journalConference article

5 Citations (Scopus)

Abstract

A general purpose neuro-image processor architecture based on the Multiplexing Neuron Array(MNA) is proposed. It is a pseudo parallel matrix-vector multiplier. The proposed architecture can process not only vector signals but two dimensional signals as well, so various types of neural networks or image processing paradigm can be implemented without any hardware modification. The major advantage is an unlimited expansion capability due to simple routings between processing elements. The architecture can be implemented with any discrete-time system, i.e., digital system, Switched-Capacitor, Switched-Current. For illustration, a multilayer perceptron network and a recurrent network are implemented with fabricated Switched-Capacitor CMOS chips.

Original languageEnglish
Pages (from-to)495-498
Number of pages4
JournalProceedings - IEEE International Symposium on Circuits and Systems
Volume3
Publication statusPublished - 1996 Jan 1

Fingerprint

Capacitors
Multilayer neural networks
Multiplexing
Neurons
Image processing
Neural networks
Hardware
Processing

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

@article{9a1bfa3995494710822ba3f351a50381,
title = "General purpose neuro-image processor architecture",
abstract = "A general purpose neuro-image processor architecture based on the Multiplexing Neuron Array(MNA) is proposed. It is a pseudo parallel matrix-vector multiplier. The proposed architecture can process not only vector signals but two dimensional signals as well, so various types of neural networks or image processing paradigm can be implemented without any hardware modification. The major advantage is an unlimited expansion capability due to simple routings between processing elements. The architecture can be implemented with any discrete-time system, i.e., digital system, Switched-Capacitor, Switched-Current. For illustration, a multilayer perceptron network and a recurrent network are implemented with fabricated Switched-Capacitor CMOS chips.",
author = "Gunhee Han and Edgar Sanchez-Sinencio",
year = "1996",
month = "1",
day = "1",
language = "English",
volume = "3",
pages = "495--498",
journal = "Proceedings - IEEE International Symposium on Circuits and Systems",
issn = "0271-4310",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

General purpose neuro-image processor architecture. / Han, Gunhee; Sanchez-Sinencio, Edgar.

In: Proceedings - IEEE International Symposium on Circuits and Systems, Vol. 3, 01.01.1996, p. 495-498.

Research output: Contribution to journalConference article

TY - JOUR

T1 - General purpose neuro-image processor architecture

AU - Han, Gunhee

AU - Sanchez-Sinencio, Edgar

PY - 1996/1/1

Y1 - 1996/1/1

N2 - A general purpose neuro-image processor architecture based on the Multiplexing Neuron Array(MNA) is proposed. It is a pseudo parallel matrix-vector multiplier. The proposed architecture can process not only vector signals but two dimensional signals as well, so various types of neural networks or image processing paradigm can be implemented without any hardware modification. The major advantage is an unlimited expansion capability due to simple routings between processing elements. The architecture can be implemented with any discrete-time system, i.e., digital system, Switched-Capacitor, Switched-Current. For illustration, a multilayer perceptron network and a recurrent network are implemented with fabricated Switched-Capacitor CMOS chips.

AB - A general purpose neuro-image processor architecture based on the Multiplexing Neuron Array(MNA) is proposed. It is a pseudo parallel matrix-vector multiplier. The proposed architecture can process not only vector signals but two dimensional signals as well, so various types of neural networks or image processing paradigm can be implemented without any hardware modification. The major advantage is an unlimited expansion capability due to simple routings between processing elements. The architecture can be implemented with any discrete-time system, i.e., digital system, Switched-Capacitor, Switched-Current. For illustration, a multilayer perceptron network and a recurrent network are implemented with fabricated Switched-Capacitor CMOS chips.

UR - http://www.scopus.com/inward/record.url?scp=0029708018&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029708018&partnerID=8YFLogxK

M3 - Conference article

VL - 3

SP - 495

EP - 498

JO - Proceedings - IEEE International Symposium on Circuits and Systems

JF - Proceedings - IEEE International Symposium on Circuits and Systems

SN - 0271-4310

ER -