Automatic segmentation of HeLa cell images

TL;DR

This paper explores various algorithms for segmenting HeLa cell images, improving their accuracy and robustness across different cell types, and discusses the process and challenges involved.

Contribution

It introduces an improved method for cell segmentation in digital images, detailing algorithm combinations and processing steps tailored for diverse cell images.

Findings

Effective segmentation across multiple cell types

Algorithm combination enhances segmentation quality

Provides insights into processing order and future directions

Abstract

In this work, the possibilities for segmentation of cells from their background and each other in digital image were tested, combined and improoved. Lot of images with young, adult and mixture cells were able to prove the quality of described algorithms. Proper segmentation is one of the main task of image analysis and steps order differ from work to work, depending on input images. Reply for biologicaly given question was looking for in this work, including filtration, details emphasizing, segmentation and sphericity computing. Order of algorithms and way to searching for them was also described. Some questions and ideas for further work were mentioned in the conclusion part.