Multi-channel spatio-temporal topographic processing for visual search and navigation

István Szatmári, Dávid Bálya, Gergely Tímár, Csaba Rekeczky, T. Roska

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

6 Citations (Scopus)

Abstract

In this paper a biologically motivated image flow processing mechanism is presented for visual exploration systems. The intention of this multi-channel topographic approach was to produce decision maps for salient feature localization and identification. As a unique biological study has recently confirmed, mammalian visual systems process the world through a set of separate parallel channels and these representations are embodied in a stack of strata in the retina. Beyond reflecting the biological motivations, our main goal was to create an efficient algorithmic framework for real-life visual search and navigation experiments. In the course of this design the retinotopic processing scheme is embedded in an analogic Cellular Neural Network (CNN) algorithm where image flow is analyzed by temporal, spatial and spatio-temporal filters. The output of these sub-channels is then combined in a programmable configuration to form the new channel responses. In the core of the algorithm crisp or fuzzy logic strategies define the global channel interaction and result in a unique binary image flow. This processing mechanism of the algorithmic framework and the hardware architecture of the system is presented along with experimental ACE4k CNN chip results for several video flows recorded in flying vehicles.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsA. Rodriguez - Vazquez, D. Abbott, R. Carmona
Pages297-306
Number of pages10
Volume5119
DOIs
Publication statusPublished - 2003
EventBioengineered and Bioinspired Systems - Maspalomas, Gran Canaria, Spain
Duration: May 19 2003May 21 2003

Other

OtherBioengineered and Bioinspired Systems
CountrySpain
CityMaspalomas, Gran Canaria
Period5/19/035/21/03

Fingerprint

navigation
Cellular neural networks
Navigation
Processing
Binary images
Fuzzy logic
retina
Hardware
strata
logic
vehicles
hardware
chips
flight
filters
Experiments
output
configurations
interactions

Keywords

  • Cellular neural network
  • Optical navigation
  • Spatio-temporal filters
  • Topographic processing
  • Visual terrain exploration

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Szatmári, I., Bálya, D., Tímár, G., Rekeczky, C., & Roska, T. (2003). Multi-channel spatio-temporal topographic processing for visual search and navigation. In A. Rodriguez - Vazquez, D. Abbott, & R. Carmona (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5119, pp. 297-306) https://doi.org/10.1117/12.501159

Multi-channel spatio-temporal topographic processing for visual search and navigation. / Szatmári, István; Bálya, Dávid; Tímár, Gergely; Rekeczky, Csaba; Roska, T.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / A. Rodriguez - Vazquez; D. Abbott; R. Carmona. Vol. 5119 2003. p. 297-306.

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

Szatmári, I, Bálya, D, Tímár, G, Rekeczky, C & Roska, T 2003, Multi-channel spatio-temporal topographic processing for visual search and navigation. in A Rodriguez - Vazquez, D Abbott & R Carmona (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5119, pp. 297-306, Bioengineered and Bioinspired Systems, Maspalomas, Gran Canaria, Spain, 5/19/03. https://doi.org/10.1117/12.501159
Szatmári I, Bálya D, Tímár G, Rekeczky C, Roska T. Multi-channel spatio-temporal topographic processing for visual search and navigation. In Rodriguez - Vazquez A, Abbott D, Carmona R, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5119. 2003. p. 297-306 https://doi.org/10.1117/12.501159
Szatmári, István ; Bálya, Dávid ; Tímár, Gergely ; Rekeczky, Csaba ; Roska, T. / Multi-channel spatio-temporal topographic processing for visual search and navigation. Proceedings of SPIE - The International Society for Optical Engineering. editor / A. Rodriguez - Vazquez ; D. Abbott ; R. Carmona. Vol. 5119 2003. pp. 297-306
@inproceedings{7447aa41836e421bbf7b0b49706456f8,
title = "Multi-channel spatio-temporal topographic processing for visual search and navigation",
abstract = "In this paper a biologically motivated image flow processing mechanism is presented for visual exploration systems. The intention of this multi-channel topographic approach was to produce decision maps for salient feature localization and identification. As a unique biological study has recently confirmed, mammalian visual systems process the world through a set of separate parallel channels and these representations are embodied in a stack of strata in the retina. Beyond reflecting the biological motivations, our main goal was to create an efficient algorithmic framework for real-life visual search and navigation experiments. In the course of this design the retinotopic processing scheme is embedded in an analogic Cellular Neural Network (CNN) algorithm where image flow is analyzed by temporal, spatial and spatio-temporal filters. The output of these sub-channels is then combined in a programmable configuration to form the new channel responses. In the core of the algorithm crisp or fuzzy logic strategies define the global channel interaction and result in a unique binary image flow. This processing mechanism of the algorithmic framework and the hardware architecture of the system is presented along with experimental ACE4k CNN chip results for several video flows recorded in flying vehicles.",
keywords = "Cellular neural network, Optical navigation, Spatio-temporal filters, Topographic processing, Visual terrain exploration",
author = "Istv{\'a}n Szatm{\'a}ri and D{\'a}vid B{\'a}lya and Gergely T{\'i}m{\'a}r and Csaba Rekeczky and T. Roska",
year = "2003",
doi = "10.1117/12.501159",
language = "English",
volume = "5119",
pages = "297--306",
editor = "{Rodriguez - Vazquez}, A. and D. Abbott and R. Carmona",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Multi-channel spatio-temporal topographic processing for visual search and navigation

AU - Szatmári, István

AU - Bálya, Dávid

AU - Tímár, Gergely

AU - Rekeczky, Csaba

AU - Roska, T.

PY - 2003

Y1 - 2003

N2 - In this paper a biologically motivated image flow processing mechanism is presented for visual exploration systems. The intention of this multi-channel topographic approach was to produce decision maps for salient feature localization and identification. As a unique biological study has recently confirmed, mammalian visual systems process the world through a set of separate parallel channels and these representations are embodied in a stack of strata in the retina. Beyond reflecting the biological motivations, our main goal was to create an efficient algorithmic framework for real-life visual search and navigation experiments. In the course of this design the retinotopic processing scheme is embedded in an analogic Cellular Neural Network (CNN) algorithm where image flow is analyzed by temporal, spatial and spatio-temporal filters. The output of these sub-channels is then combined in a programmable configuration to form the new channel responses. In the core of the algorithm crisp or fuzzy logic strategies define the global channel interaction and result in a unique binary image flow. This processing mechanism of the algorithmic framework and the hardware architecture of the system is presented along with experimental ACE4k CNN chip results for several video flows recorded in flying vehicles.

AB - In this paper a biologically motivated image flow processing mechanism is presented for visual exploration systems. The intention of this multi-channel topographic approach was to produce decision maps for salient feature localization and identification. As a unique biological study has recently confirmed, mammalian visual systems process the world through a set of separate parallel channels and these representations are embodied in a stack of strata in the retina. Beyond reflecting the biological motivations, our main goal was to create an efficient algorithmic framework for real-life visual search and navigation experiments. In the course of this design the retinotopic processing scheme is embedded in an analogic Cellular Neural Network (CNN) algorithm where image flow is analyzed by temporal, spatial and spatio-temporal filters. The output of these sub-channels is then combined in a programmable configuration to form the new channel responses. In the core of the algorithm crisp or fuzzy logic strategies define the global channel interaction and result in a unique binary image flow. This processing mechanism of the algorithmic framework and the hardware architecture of the system is presented along with experimental ACE4k CNN chip results for several video flows recorded in flying vehicles.

KW - Cellular neural network

KW - Optical navigation

KW - Spatio-temporal filters

KW - Topographic processing

KW - Visual terrain exploration

UR - http://www.scopus.com/inward/record.url?scp=0041827000&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041827000&partnerID=8YFLogxK

U2 - 10.1117/12.501159

DO - 10.1117/12.501159

M3 - Conference contribution

AN - SCOPUS:0041827000

VL - 5119

SP - 297

EP - 306

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Rodriguez - Vazquez, A.

A2 - Abbott, D.

A2 - Carmona, R.

ER -