Self-organizing scaling filters for image segmentation

T. Rozgonyi, T. Fomin, A. Lőrincz

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

1 Citation (Scopus)

Abstract

The feasibility of using neural networks to create spatial filters of different sizes for solving fast image segmentation tasks in parallel was evaluated. A mathematical model consisting of several competing neural units and having no adjustable parameter was used in the study. Results indicate that the network has a winner-take-all learning mechanism and a self-organizing learning rule which can be used to develop overlapping circular filters of local Gaussian character.

Original languageEnglish
Title of host publicationIEEE International Conference on Neural Networks - Conference Proceedings
PublisherIEEE
Pages4380-4383
Number of pages4
Volume7
Publication statusPublished - 1994
EventProceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) - Orlando, FL, USA
Duration: Jun 27 1994Jun 29 1994

Other

OtherProceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7)
CityOrlando, FL, USA
Period6/27/946/29/94

Fingerprint

Image segmentation
Mathematical models
Neural networks

ASJC Scopus subject areas

  • Software

Cite this

Rozgonyi, T., Fomin, T., & Lőrincz, A. (1994). Self-organizing scaling filters for image segmentation. In IEEE International Conference on Neural Networks - Conference Proceedings (Vol. 7, pp. 4380-4383). IEEE.

Self-organizing scaling filters for image segmentation. / Rozgonyi, T.; Fomin, T.; Lőrincz, A.

IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 7 IEEE, 1994. p. 4380-4383.

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

Rozgonyi, T, Fomin, T & Lőrincz, A 1994, Self-organizing scaling filters for image segmentation. in IEEE International Conference on Neural Networks - Conference Proceedings. vol. 7, IEEE, pp. 4380-4383, Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7), Orlando, FL, USA, 6/27/94.
Rozgonyi T, Fomin T, Lőrincz A. Self-organizing scaling filters for image segmentation. In IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 7. IEEE. 1994. p. 4380-4383
Rozgonyi, T. ; Fomin, T. ; Lőrincz, A. / Self-organizing scaling filters for image segmentation. IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 7 IEEE, 1994. pp. 4380-4383
@inproceedings{3dd9e1d52def4044adf5dfd56f8e1621,
title = "Self-organizing scaling filters for image segmentation",
abstract = "The feasibility of using neural networks to create spatial filters of different sizes for solving fast image segmentation tasks in parallel was evaluated. A mathematical model consisting of several competing neural units and having no adjustable parameter was used in the study. Results indicate that the network has a winner-take-all learning mechanism and a self-organizing learning rule which can be used to develop overlapping circular filters of local Gaussian character.",
author = "T. Rozgonyi and T. Fomin and A. Lőrincz",
year = "1994",
language = "English",
volume = "7",
pages = "4380--4383",
booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Self-organizing scaling filters for image segmentation

AU - Rozgonyi, T.

AU - Fomin, T.

AU - Lőrincz, A.

PY - 1994

Y1 - 1994

N2 - The feasibility of using neural networks to create spatial filters of different sizes for solving fast image segmentation tasks in parallel was evaluated. A mathematical model consisting of several competing neural units and having no adjustable parameter was used in the study. Results indicate that the network has a winner-take-all learning mechanism and a self-organizing learning rule which can be used to develop overlapping circular filters of local Gaussian character.

AB - The feasibility of using neural networks to create spatial filters of different sizes for solving fast image segmentation tasks in parallel was evaluated. A mathematical model consisting of several competing neural units and having no adjustable parameter was used in the study. Results indicate that the network has a winner-take-all learning mechanism and a self-organizing learning rule which can be used to develop overlapping circular filters of local Gaussian character.

UR - http://www.scopus.com/inward/record.url?scp=0028710575&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028710575&partnerID=8YFLogxK

M3 - Conference contribution

VL - 7

SP - 4380

EP - 4383

BT - IEEE International Conference on Neural Networks - Conference Proceedings

PB - IEEE

ER -