Open AccessPossibilities of automatic assessment of the vitreous nature of wheat and linear characteristics of grain (seeds) by digital image analysis
Author(s) -
T. S. Rutkovskaya,
R. Yu. Antonov,
Георги Петров
Publication year - 2020
Publication title -
agrarnaâ nauka
Language(s) - English
Resource type - Journals
eISSN - 2686-701X
pISSN - 0869-8155
DOI - 10.32634/0869-8155-2020-338-5-87-90
Subject(s) - digital image analysis , organoleptic , stability (learning theory) , relevance (law) , linear relationship , digital image , sample (material) , digital image processing , image processing , computer science , mathematics , wheat grain , artificial intelligence , computer vision , statistics , image (mathematics) , pattern recognition (psychology) , biological system , food science , machine learning , chemistry , biology , chromatography , agronomy , political science , law
Relevance and methods. The article considers the possibility of using the analysis of digital images for a comprehensive assessment of the physical characteristics of grain: vitreous and linear dimensions. Results. A comparative characteristic of instrumental and organoleptic methods for determining these indicators by the following criteria is given: stability of results, speed of measurements and data processing. An algorithm has been developed that combines the ability to programmatically determine glassiness and linear dimensions of wheat grains based on their digital images. A relationship was found between an increase in the vitreous nature of the sample and an increase in the stability of the results.