Author(s): Balázs VÍGH

Title: KÉPSZEGMENTÁLÓ ALGORITMUSOK ÉRTÉKELÉSE PENÉSZKOLÓNIÁK NÖVEKEDÉSÉNEK ELEMZÉSÉRE

Source: Š. Gubo, E. Kahler Korcsmáros, Sz. Svitek, J. Tóth (eds.): 17th International Scientific Conference of J. Selye University. Sessions of the Faculty of Economics and Informatics. Conference Proceedings

ISBN: 978-80-8122-538-3

DOI: https://doi.org/10.36007/5383.2026.65

Publisher: J. Selye University, Komárno, Slovakia

PY, pages: 2025, 65-75

Published on-line: 2026

Language: hu

Abstract: The aim of this thesis is to present an image processing pipeline for analyzing the growth of mold colonies on agar plates. The developed program has a modular structure and includes preprocessing steps, different segmentation algorithms, and an evaluation module. The segmentation masks were compared with annotations made using the Ilastik software, and the evaluation was carried out with standard metrics. The methodology was tested on a limited number of images, with the main purpose of demonstrating the processing steps and presenting an expandable, modular program design. The proposed pipeline can be used to compare the performance of different segmentation algorithms and may serve as a basis for later statistical studies on larger datasets.

Keywords: Algorithm evaluation, Image processing pipeline, Image segmentation, Modular software, Mold colony growth

Fulltext (PDF)