Real-time analog VLSI sensors for 2-D direction of motion

Rainer A. Deutschmarm, Charles M Higgins, Christof Koch

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Optical flow fields are a primary source of information about the visual scene in technical and biological systems. In a step towards a system for real time scene analysis we have developed two new algorithms for the parallel computation of the direction of motion field in 2-D. We have successfully implemented these algorithms in analog VLSI hardware such that all processing is performed in the focal plane of the imager. We present data from 12 x 13 and 15 x 14 test arrays and show that the sensors are reliable for over one order of magnitude in velocity and down to 20% contrast. The algorithms allow for a very small pixel size (110µm x 110µm in а 1.2/ітп CMOS process) and thus high resolution arrays are feasible.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages1163-1168
Number of pages6
Volume1327
ISBN (Print)3540636315, 9783540636311
StatePublished - 1997
Externally publishedYes
Event7th International Conference on Artificial Neural Networks, ICANN 1997 - Lausanne, Switzerland
Duration: Oct 8 1997Oct 10 1997

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1327
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other7th International Conference on Artificial Neural Networks, ICANN 1997
CountrySwitzerland
CityLausanne
Period10/8/9710/10/97

Fingerprint

Analogue
Real-time
Sensor
Motion
Sensors
Scene Analysis
Optical flows
Optical Flow
Parallel Computation
Biological systems
Focal Plane
Imager
Biological Systems
Image sensors
Flow Field
Flow fields
High Resolution
Pixel
Pixels
Hardware

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Deutschmarm, R. A., Higgins, C. M., & Koch, C. (1997). Real-time analog VLSI sensors for 2-D direction of motion. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1327, pp. 1163-1168). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1327). Springer Verlag.

Real-time analog VLSI sensors for 2-D direction of motion. / Deutschmarm, Rainer A.; Higgins, Charles M; Koch, Christof.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1327 Springer Verlag, 1997. p. 1163-1168 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1327).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Deutschmarm, RA, Higgins, CM & Koch, C 1997, Real-time analog VLSI sensors for 2-D direction of motion. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 1327, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1327, Springer Verlag, pp. 1163-1168, 7th International Conference on Artificial Neural Networks, ICANN 1997, Lausanne, Switzerland, 10/8/97.
Deutschmarm RA, Higgins CM, Koch C. Real-time analog VLSI sensors for 2-D direction of motion. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1327. Springer Verlag. 1997. p. 1163-1168. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Deutschmarm, Rainer A. ; Higgins, Charles M ; Koch, Christof. / Real-time analog VLSI sensors for 2-D direction of motion. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1327 Springer Verlag, 1997. pp. 1163-1168 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{93197c07356b4e4782f37e754f9b5356,
title = "Real-time analog VLSI sensors for 2-D direction of motion",
abstract = "Optical flow fields are a primary source of information about the visual scene in technical and biological systems. In a step towards a system for real time scene analysis we have developed two new algorithms for the parallel computation of the direction of motion field in 2-D. We have successfully implemented these algorithms in analog VLSI hardware such that all processing is performed in the focal plane of the imager. We present data from 12 x 13 and 15 x 14 test arrays and show that the sensors are reliable for over one order of magnitude in velocity and down to 20{\%} contrast. The algorithms allow for a very small pixel size (110µm x 110µm in а 1.2/ітп CMOS process) and thus high resolution arrays are feasible.",
author = "Deutschmarm, {Rainer A.} and Higgins, {Charles M} and Christof Koch",
year = "1997",
language = "English (US)",
isbn = "3540636315",
volume = "1327",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "1163--1168",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Real-time analog VLSI sensors for 2-D direction of motion

AU - Deutschmarm, Rainer A.

AU - Higgins, Charles M

AU - Koch, Christof

PY - 1997

Y1 - 1997

N2 - Optical flow fields are a primary source of information about the visual scene in technical and biological systems. In a step towards a system for real time scene analysis we have developed two new algorithms for the parallel computation of the direction of motion field in 2-D. We have successfully implemented these algorithms in analog VLSI hardware such that all processing is performed in the focal plane of the imager. We present data from 12 x 13 and 15 x 14 test arrays and show that the sensors are reliable for over one order of magnitude in velocity and down to 20% contrast. The algorithms allow for a very small pixel size (110µm x 110µm in а 1.2/ітп CMOS process) and thus high resolution arrays are feasible.

AB - Optical flow fields are a primary source of information about the visual scene in technical and biological systems. In a step towards a system for real time scene analysis we have developed two new algorithms for the parallel computation of the direction of motion field in 2-D. We have successfully implemented these algorithms in analog VLSI hardware such that all processing is performed in the focal plane of the imager. We present data from 12 x 13 and 15 x 14 test arrays and show that the sensors are reliable for over one order of magnitude in velocity and down to 20% contrast. The algorithms allow for a very small pixel size (110µm x 110µm in а 1.2/ітп CMOS process) and thus high resolution arrays are feasible.

UR - http://www.scopus.com/inward/record.url?scp=84908448541&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908448541&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540636315

SN - 9783540636311

VL - 1327

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 1163

EP - 1168

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -