Neural network based adaptive image segmentation

Shen Dinggang, Qi Feihu

Research output: Contribution to journalConference articlepeer-review

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
Pages (from-to)1035-1038
Number of pages4
JournalNational Conference Publication - Institution of Engineers, Australia
Volume2
Issue number94 /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

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Neural network based adaptive image segmentation'. Together they form a unique fingerprint.

Cite this