Application of k-means clustering and histogram analysis to automate preprocessing of images of discomycetes obtained in the habitat D. A. Filimonova, I. G. Vorob’eva, A. Yu. Filimonov
Material type:
ArticleContent type: Текст Media type: электронный Other title: Применение кластеризации k-means и анализа гистограмм для автоматизации предварительной обработки изображений дискомицетов, полученных в среде обитания [Parallel title]Subject(s): кластеризация | компьютерное зрение | машинное обучение | дискомицеты | биоразнообразиеGenre/Form: статьи в журналах Online resources: Click here to access online
In:
Вестник Томского государственного университета. Управление, вычислительная техника и информатика № 63. С. 111-117Abstract: The study of biological diversity requires a thorough inventory of all groups of organisms, including destructors, among which fungi play a significant role. Discomycetes, a group of orders of fungi of the Ascomycota phylum, require close attention from researchers due to their low level of knowledge. The paper proposes an approach to automating the process of inventory of representatives of this group of orders and presents a prototype of a software package that allows one to identify the presence of fruit bodies of discomycetes in photographs taken in the natural habitat. A feature of the proposed solution is the application of the k-means clustering method, the use of scaled histograms to determine the presence of an image of the fruit body of Discomycetes in this image, and the prospects for using this tool in machine learning are described using neural networks.
Библиогр.: 16 назв.
The study of biological diversity requires a thorough inventory of all groups of organisms, including destructors, among which fungi play a significant role. Discomycetes, a group of orders of fungi of the Ascomycota phylum, require close attention from researchers due to their low level of knowledge. The paper proposes an approach to automating the process of inventory of representatives of this group of orders and presents a prototype of a software package that allows one to identify the presence of fruit bodies of discomycetes in photographs taken in the natural habitat. A feature of the proposed solution is the application of the k-means clustering method, the use of scaled histograms to determine the presence of an image of the fruit body of Discomycetes in this image, and the prospects for using this tool in machine learning are described using neural networks.
There are no comments on this title.