Morphological image denoising and random-walks spot detection in microarry images

A. Mastrogianni, E. Dermatas, A. Bezerianos

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

1 Scopus citations

Abstract

In this paper, a novel method for noise reduction and spot detection in microarray images, based on mathematical morphology and Random-Walks model is presented and evaluated. Initially, an automatic method detects the subarrays and a morphological denoising method is used to reduce artifacts at each individual subarray. In the enhanced images, a novel segmentation algorithm based on Random-Walker algorithm (RWA) derives the spot positions. The proposed method is validated in two different databases of real distorted microarray images, giving quite satisfying results both in noise elimination and spot detection accuracy.

Original languageEnglish (US)
Title of host publicationProceedings of the 9th International Workshop on Mathematical Methods in Scattering Theory and Biomedical Engineering
Subtitle of host publicationAdvanced Topics in Scattering and Biomedical Engineering
PublisherWorld Scientific Publishing Co. Pte Ltd
Pages143-150
Number of pages8
ISBN (Print)9814322024, 9789814322027
DOIs
StatePublished - 2010
Externally publishedYes
Event2009 9th International Workshop on Mathematical Methods in Scattering Theory and Biomedical Engineering - Patras, Greece
Duration: Oct 9 2009Oct 11 2009

Publication series

NameProceedings of the 9th International Workshop on Mathematical Methods in Scattering Theory and Biomedical Engineering: Advanced Topics in Scattering and Biomedical Engineering

Other

Other2009 9th International Workshop on Mathematical Methods in Scattering Theory and Biomedical Engineering
Country/TerritoryGreece
CityPatras
Period10/9/0910/11/09

ASJC Scopus subject areas

  • Biomedical Engineering
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Morphological image denoising and random-walks spot detection in microarry images'. Together they form a unique fingerprint.

Cite this