Image segmentation and particles classification using texture analysis method

Atteya M., Salem M., Hegazy, Doaa, Roushdy M.,


� 2016, Sociedade Brasileira de Engenharia Biomedica. All rights reserved. Introduction: Ingredients of oily fish include a large amount of polyunsaturated fatty acids, which are important elements in various metabolic processes of humans, and have also been used to prevent diseases. However, in an attempt to reduce cost, recent developments are starting a replace the ingredients of fish oil with products of microalgae, that also produce polyunsaturated fatty acids. To do so, it is important to closely monitor morphological changes in algae cells and monitor their age in order to achieve the best results. This paper aims to describe an advanced vision-based system to automatically detect, classify, and track the organic cells using a recently developed SOPAT-System (Smart On-line Particle Analysis Technology), a photo-optical image acquisition device combined with innovative image analysis software. Methods: The proposed method includes image de-noising, binarization and Enhancement, as well as object recognition, localization and classification based on the analysis of particles’ size and texture. Results: The methods allowed for correctly computing cell’s size for each particle separately. By computing an area histogram for the input images (1h, 18h, and 42h), the variation could be observed showing a clear increase in cell. Conclusion: The proposed method allows for algae particles to be correctly identified with accuracies up to 99% and classified correctly with accuracies up to 100%.

Other data

Title Image segmentation and particles classification using texture analysis method
Authors Atteya M. ; Salem M. ; Hegazy, Doaa ; Roushdy M. 
Issue Date 1-Jul-2016
Journal Revista Brasileira de Engenharia Biomedica 
Scopus ID 2-s2.0-84994140758

Recommend this item

Similar Items from Core Recommender Database

Google ScholarTM


Citations 4 in scopus
views 24 in Shams Scholar

Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.