Modular multi-chip neuromorphic architecture for real-time visual motion processing

Charles M Higgins, Christof Koch

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

The extent of pixel-parallel focal plane image processing is limited by pixel area and imager fill factor. In this paper, we describe a novel multi-chip neuromorphic VLSI visual motion processing system which combines analog circuitry with an asynchronous digital interchip communications protocol to allow more complex pixel-parallel motion processing than is possible in the focal plane. This multi-chip system retains the primary advantages of focal plane neuromorphic image processors: low-power consumption, continuous-time operation, and small size. The two basic VLSI building blocks are a photosensitive sender chip which incorporates a 2D imager array and transmits the position of moving spatial edges, and a receiver chip which computes a 2D optical flow vector field from the edge information. The elementary two-chip motion processing system consisting of a single sender and receiver is first characterized. Subsequently, two three-chip motion processing systems are described. The first three-chip system uses two sender chips to compute the presence of motion only at a particular stereoscopic depth from the imagers. The second three-chip system uses two receivers to simultaneously compute a linear and polar topographic mapping of the image plane, resulting in information about image translation, rotation, and expansion. These three-chip systems demonstrate the modularity and flexibility of the multi-chip neuromorphic approach.

Original languageEnglish (US)
Pages (from-to)195-211
Number of pages17
JournalAnalog Integrated Circuits and Signal Processing
Volume24
Issue number3
DOIs
StatePublished - Sep 2000

Fingerprint

Image sensors
Pixels
Processing
Optical flows
Image processing
Electric power utilization
Network protocols

ASJC Scopus subject areas

  • Hardware and Architecture
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Modular multi-chip neuromorphic architecture for real-time visual motion processing. / Higgins, Charles M; Koch, Christof.

In: Analog Integrated Circuits and Signal Processing, Vol. 24, No. 3, 09.2000, p. 195-211.

Research output: Contribution to journalArticle

@article{94b6bf3845cc430fb0ccd8a8fec9bbe4,
title = "Modular multi-chip neuromorphic architecture for real-time visual motion processing",
abstract = "The extent of pixel-parallel focal plane image processing is limited by pixel area and imager fill factor. In this paper, we describe a novel multi-chip neuromorphic VLSI visual motion processing system which combines analog circuitry with an asynchronous digital interchip communications protocol to allow more complex pixel-parallel motion processing than is possible in the focal plane. This multi-chip system retains the primary advantages of focal plane neuromorphic image processors: low-power consumption, continuous-time operation, and small size. The two basic VLSI building blocks are a photosensitive sender chip which incorporates a 2D imager array and transmits the position of moving spatial edges, and a receiver chip which computes a 2D optical flow vector field from the edge information. The elementary two-chip motion processing system consisting of a single sender and receiver is first characterized. Subsequently, two three-chip motion processing systems are described. The first three-chip system uses two sender chips to compute the presence of motion only at a particular stereoscopic depth from the imagers. The second three-chip system uses two receivers to simultaneously compute a linear and polar topographic mapping of the image plane, resulting in information about image translation, rotation, and expansion. These three-chip systems demonstrate the modularity and flexibility of the multi-chip neuromorphic approach.",
author = "Higgins, {Charles M} and Christof Koch",
year = "2000",
month = "9",
doi = "10.1023/A:1008309524326",
language = "English (US)",
volume = "24",
pages = "195--211",
journal = "Analog Integrated Circuits and Signal Processing",
issn = "0925-1030",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Modular multi-chip neuromorphic architecture for real-time visual motion processing

AU - Higgins, Charles M

AU - Koch, Christof

PY - 2000/9

Y1 - 2000/9

N2 - The extent of pixel-parallel focal plane image processing is limited by pixel area and imager fill factor. In this paper, we describe a novel multi-chip neuromorphic VLSI visual motion processing system which combines analog circuitry with an asynchronous digital interchip communications protocol to allow more complex pixel-parallel motion processing than is possible in the focal plane. This multi-chip system retains the primary advantages of focal plane neuromorphic image processors: low-power consumption, continuous-time operation, and small size. The two basic VLSI building blocks are a photosensitive sender chip which incorporates a 2D imager array and transmits the position of moving spatial edges, and a receiver chip which computes a 2D optical flow vector field from the edge information. The elementary two-chip motion processing system consisting of a single sender and receiver is first characterized. Subsequently, two three-chip motion processing systems are described. The first three-chip system uses two sender chips to compute the presence of motion only at a particular stereoscopic depth from the imagers. The second three-chip system uses two receivers to simultaneously compute a linear and polar topographic mapping of the image plane, resulting in information about image translation, rotation, and expansion. These three-chip systems demonstrate the modularity and flexibility of the multi-chip neuromorphic approach.

AB - The extent of pixel-parallel focal plane image processing is limited by pixel area and imager fill factor. In this paper, we describe a novel multi-chip neuromorphic VLSI visual motion processing system which combines analog circuitry with an asynchronous digital interchip communications protocol to allow more complex pixel-parallel motion processing than is possible in the focal plane. This multi-chip system retains the primary advantages of focal plane neuromorphic image processors: low-power consumption, continuous-time operation, and small size. The two basic VLSI building blocks are a photosensitive sender chip which incorporates a 2D imager array and transmits the position of moving spatial edges, and a receiver chip which computes a 2D optical flow vector field from the edge information. The elementary two-chip motion processing system consisting of a single sender and receiver is first characterized. Subsequently, two three-chip motion processing systems are described. The first three-chip system uses two sender chips to compute the presence of motion only at a particular stereoscopic depth from the imagers. The second three-chip system uses two receivers to simultaneously compute a linear and polar topographic mapping of the image plane, resulting in information about image translation, rotation, and expansion. These three-chip systems demonstrate the modularity and flexibility of the multi-chip neuromorphic approach.

UR - http://www.scopus.com/inward/record.url?scp=0034271811&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034271811&partnerID=8YFLogxK

U2 - 10.1023/A:1008309524326

DO - 10.1023/A:1008309524326

M3 - Article

AN - SCOPUS:0034271811

VL - 24

SP - 195

EP - 211

JO - Analog Integrated Circuits and Signal Processing

JF - Analog Integrated Circuits and Signal Processing

SN - 0925-1030

IS - 3

ER -