Programmable optical CNN implementation based on the template pixels' angular coding

S. Tokés, L. Orzó, T. Roska

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

Abstract

Within the programmable opto-electronic analogic computer (POAC) framework a new, feed forward only optical CNN-UM implementation has been introduced. It is grounded on an innovative semi-incoherent optical correlator architecture. Angular coding of the template pixels determines the operation of this optical CNN implementation, therefore it is real time and flexibly programmable. We have demonstrated its feasibility and operation by an experimental setup. Our correlator architecture makes it possible to execute algorithms real time, which cannot be done by any other existing optical correlator so far. Our architecture unifies the advantages of coherent and incoherent optical correlators, provides a more robust frame and avoids their main hindrances. In the POAC framework the resulting correlogram is measured by a programmable adaptive sensor array, a special visual CNN-UM chip. So, local parallel programs fulfill both the necessary pre and post processing with the required adaptive thresholding. However, because of the limited resolution of available visual CNN chips (28x28), all-optical optical pre- and post-processing will be used, as well.

Original languageEnglish
Title of host publicationProceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pagesxxii-xxiii
Volume2002-January
ISBN (Print)981238121X
DOIs
Publication statusPublished - 2002
Event7th IEEE International Workshop on Cellular Neural Networks and their Applications, CNNA 2002 - Frankfurt, Germany
Duration: Jul 22 2002Jul 24 2002

Other

Other7th IEEE International Workshop on Cellular Neural Networks and their Applications, CNNA 2002
CountryGermany
CityFrankfurt
Period7/22/027/24/02

Fingerprint

Correlators
Template
Coding
Pixel
Pixels
Correlator
Processing
Optoelectronics
Post-processing
Sensor arrays
Preprocessing
Chip
Adaptive Thresholding
Sensor Array
Parallel Programs
Feedforward
Necessary
Architecture

Keywords

  • Adaptive arrays
  • Cellular neural networks
  • Computer architecture
  • Correlators
  • Feeds
  • Optical computing
  • Optical sensors
  • Robustness
  • Semiconductor device measurement
  • Sensor arrays

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Electrical and Electronic Engineering
  • Modelling and Simulation

Cite this

Tokés, S., Orzó, L., & Roska, T. (2002). Programmable optical CNN implementation based on the template pixels' angular coding. In Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications (Vol. 2002-January, pp. xxii-xxiii). [1035027] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CNNA.2002.1035027

Programmable optical CNN implementation based on the template pixels' angular coding. / Tokés, S.; Orzó, L.; Roska, T.

Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications. Vol. 2002-January Institute of Electrical and Electronics Engineers Inc., 2002. p. xxii-xxiii 1035027.

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

Tokés, S, Orzó, L & Roska, T 2002, Programmable optical CNN implementation based on the template pixels' angular coding. in Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications. vol. 2002-January, 1035027, Institute of Electrical and Electronics Engineers Inc., pp. xxii-xxiii, 7th IEEE International Workshop on Cellular Neural Networks and their Applications, CNNA 2002, Frankfurt, Germany, 7/22/02. https://doi.org/10.1109/CNNA.2002.1035027
Tokés S, Orzó L, Roska T. Programmable optical CNN implementation based on the template pixels' angular coding. In Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications. Vol. 2002-January. Institute of Electrical and Electronics Engineers Inc. 2002. p. xxii-xxiii. 1035027 https://doi.org/10.1109/CNNA.2002.1035027
Tokés, S. ; Orzó, L. ; Roska, T. / Programmable optical CNN implementation based on the template pixels' angular coding. Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications. Vol. 2002-January Institute of Electrical and Electronics Engineers Inc., 2002. pp. xxii-xxiii
@inproceedings{35b759aa1861444daa94406c6f234d8b,
title = "Programmable optical CNN implementation based on the template pixels' angular coding",
abstract = "Within the programmable opto-electronic analogic computer (POAC) framework a new, feed forward only optical CNN-UM implementation has been introduced. It is grounded on an innovative semi-incoherent optical correlator architecture. Angular coding of the template pixels determines the operation of this optical CNN implementation, therefore it is real time and flexibly programmable. We have demonstrated its feasibility and operation by an experimental setup. Our correlator architecture makes it possible to execute algorithms real time, which cannot be done by any other existing optical correlator so far. Our architecture unifies the advantages of coherent and incoherent optical correlators, provides a more robust frame and avoids their main hindrances. In the POAC framework the resulting correlogram is measured by a programmable adaptive sensor array, a special visual CNN-UM chip. So, local parallel programs fulfill both the necessary pre and post processing with the required adaptive thresholding. However, because of the limited resolution of available visual CNN chips (28x28), all-optical optical pre- and post-processing will be used, as well.",
keywords = "Adaptive arrays, Cellular neural networks, Computer architecture, Correlators, Feeds, Optical computing, Optical sensors, Robustness, Semiconductor device measurement, Sensor arrays",
author = "S. Tok{\'e}s and L. Orz{\'o} and T. Roska",
year = "2002",
doi = "10.1109/CNNA.2002.1035027",
language = "English",
isbn = "981238121X",
volume = "2002-January",
pages = "xxii--xxiii",
booktitle = "Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Programmable optical CNN implementation based on the template pixels' angular coding

AU - Tokés, S.

AU - Orzó, L.

AU - Roska, T.

PY - 2002

Y1 - 2002

N2 - Within the programmable opto-electronic analogic computer (POAC) framework a new, feed forward only optical CNN-UM implementation has been introduced. It is grounded on an innovative semi-incoherent optical correlator architecture. Angular coding of the template pixels determines the operation of this optical CNN implementation, therefore it is real time and flexibly programmable. We have demonstrated its feasibility and operation by an experimental setup. Our correlator architecture makes it possible to execute algorithms real time, which cannot be done by any other existing optical correlator so far. Our architecture unifies the advantages of coherent and incoherent optical correlators, provides a more robust frame and avoids their main hindrances. In the POAC framework the resulting correlogram is measured by a programmable adaptive sensor array, a special visual CNN-UM chip. So, local parallel programs fulfill both the necessary pre and post processing with the required adaptive thresholding. However, because of the limited resolution of available visual CNN chips (28x28), all-optical optical pre- and post-processing will be used, as well.

AB - Within the programmable opto-electronic analogic computer (POAC) framework a new, feed forward only optical CNN-UM implementation has been introduced. It is grounded on an innovative semi-incoherent optical correlator architecture. Angular coding of the template pixels determines the operation of this optical CNN implementation, therefore it is real time and flexibly programmable. We have demonstrated its feasibility and operation by an experimental setup. Our correlator architecture makes it possible to execute algorithms real time, which cannot be done by any other existing optical correlator so far. Our architecture unifies the advantages of coherent and incoherent optical correlators, provides a more robust frame and avoids their main hindrances. In the POAC framework the resulting correlogram is measured by a programmable adaptive sensor array, a special visual CNN-UM chip. So, local parallel programs fulfill both the necessary pre and post processing with the required adaptive thresholding. However, because of the limited resolution of available visual CNN chips (28x28), all-optical optical pre- and post-processing will be used, as well.

KW - Adaptive arrays

KW - Cellular neural networks

KW - Computer architecture

KW - Correlators

KW - Feeds

KW - Optical computing

KW - Optical sensors

KW - Robustness

KW - Semiconductor device measurement

KW - Sensor arrays

UR - http://www.scopus.com/inward/record.url?scp=84949641007&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949641007&partnerID=8YFLogxK

U2 - 10.1109/CNNA.2002.1035027

DO - 10.1109/CNNA.2002.1035027

M3 - Conference contribution

AN - SCOPUS:84949641007

SN - 981238121X

VL - 2002-January

SP - xxii-xxiii

BT - Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications

PB - Institute of Electrical and Electronics Engineers Inc.

ER -