Neural network based adaptive image segmentation

Dinggang Shen, Qi Feihu

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

Abstract

It is difficult to separate objects from background by means of thresholding, in conditions of poor and nonuniform illumination. In this paper, we present a neuralcomputing approach for image segmentation via adaptive thresholding. The thresholding surface is calculated by interpolating the image gray levels at points where the Laplacian is high, indicating probable object edges. The interpolation in the image plane is completed by Hopfield neural network. In the experiments, we show that our method is better than the global thresholding method.

Original languageEnglish
Title of host publicationNational Conference Publication - Institution of Engineers, Australia
Editors Anon
PublisherIE Aust
Pages1035-1038
Number of pages4
Volume2
Edition94 /9
Publication statusPublished - 1994
Externally publishedYes
EventProceedings of the International Symposium on Information Theory & Its Applications 1994. Part 1 (of 2) - Sydney, Aust
Duration: 1994 Nov 201994 Nov 24

Other

OtherProceedings of the International Symposium on Information Theory & Its Applications 1994. Part 1 (of 2)
CitySydney, Aust
Period94/11/2094/11/24

Fingerprint

Hopfield neural networks
Image segmentation
Interpolation
Lighting
Neural networks
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Shen, D., & Feihu, Q. (1994). Neural network based adaptive image segmentation. In Anon (Ed.), National Conference Publication - Institution of Engineers, Australia (94 /9 ed., Vol. 2, pp. 1035-1038). IE Aust.

Neural network based adaptive image segmentation. / Shen, Dinggang; Feihu, Qi.

National Conference Publication - Institution of Engineers, Australia. ed. / Anon. Vol. 2 94 /9. ed. IE Aust, 1994. p. 1035-1038.

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

Shen, D & Feihu, Q 1994, Neural network based adaptive image segmentation. in Anon (ed.), National Conference Publication - Institution of Engineers, Australia. 94 /9 edn, vol. 2, IE Aust, pp. 1035-1038, Proceedings of the International Symposium on Information Theory & Its Applications 1994. Part 1 (of 2), Sydney, Aust, 94/11/20.
Shen D, Feihu Q. Neural network based adaptive image segmentation. In Anon, editor, National Conference Publication - Institution of Engineers, Australia. 94 /9 ed. Vol. 2. IE Aust. 1994. p. 1035-1038
Shen, Dinggang ; Feihu, Qi. / Neural network based adaptive image segmentation. National Conference Publication - Institution of Engineers, Australia. editor / Anon. Vol. 2 94 /9. ed. IE Aust, 1994. pp. 1035-1038
@inproceedings{6aa53173a2ca448088ceacf4c873f70b,
title = "Neural network based adaptive image segmentation",
abstract = "It is difficult to separate objects from background by means of thresholding, in conditions of poor and nonuniform illumination. In this paper, we present a neuralcomputing approach for image segmentation via adaptive thresholding. The thresholding surface is calculated by interpolating the image gray levels at points where the Laplacian is high, indicating probable object edges. The interpolation in the image plane is completed by Hopfield neural network. In the experiments, we show that our method is better than the global thresholding method.",
author = "Dinggang Shen and Qi Feihu",
year = "1994",
language = "English",
volume = "2",
pages = "1035--1038",
editor = "Anon",
booktitle = "National Conference Publication - Institution of Engineers, Australia",
publisher = "IE Aust",
edition = "94 /9",

}

TY - GEN

T1 - Neural network based adaptive image segmentation

AU - Shen, Dinggang

AU - Feihu, Qi

PY - 1994

Y1 - 1994

N2 - It is difficult to separate objects from background by means of thresholding, in conditions of poor and nonuniform illumination. In this paper, we present a neuralcomputing approach for image segmentation via adaptive thresholding. The thresholding surface is calculated by interpolating the image gray levels at points where the Laplacian is high, indicating probable object edges. The interpolation in the image plane is completed by Hopfield neural network. In the experiments, we show that our method is better than the global thresholding method.

AB - It is difficult to separate objects from background by means of thresholding, in conditions of poor and nonuniform illumination. In this paper, we present a neuralcomputing approach for image segmentation via adaptive thresholding. The thresholding surface is calculated by interpolating the image gray levels at points where the Laplacian is high, indicating probable object edges. The interpolation in the image plane is completed by Hopfield neural network. In the experiments, we show that our method is better than the global thresholding method.

UR - http://www.scopus.com/inward/record.url?scp=0028754705&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028754705&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0028754705

VL - 2

SP - 1035

EP - 1038

BT - National Conference Publication - Institution of Engineers, Australia

A2 - Anon, null

PB - IE Aust

ER -