Multimodal image segmentation of cellular fragmentation using edge detector and morphological operators

A. Khan, C. Weiss, B. Schweitzer, I. Hansjosten, R. Mikut, M. Reischl

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Image processing and analysis pipelines can be adapted to solve challenging biological cell segmentation problems. Occasionally, the image segmentation outcomes of images containing cells are unsatisfactory and insufficient if done alone in a single channel e.g. when staining the nucleus. An efficient method using morphological operators and edge detectors using information from an additional bright field (BF) channel is able to assign conflicting cases when segments of nuclei belonging to a single cell are detected artificially as independent objects by nuclear staining. This method is shown to improve the image segmentation accuracy of cell segmentation using the information from multiple channels in comparison to only using nuclear staining.

Original languageEnglish
Pages (from-to)S518-S521
JournalBiomedizinische Technik
Volume59
DOIs
StatePublished - 1 Oct 2014
Externally publishedYes

Fingerprint

Dive into the research topics of 'Multimodal image segmentation of cellular fragmentation using edge detector and morphological operators'. Together they form a unique fingerprint.

Cite this