Machine learning in solving problems of producing the most important bread cereals in the Kuban

The article is devoted to the main issues of the application of modern mathematical methods for solving problems related to biological objects. An excursion into the history of the emergence of machine learning methods, which today are most in demand for solving the most important tasks of the agro-...

Full description

Saved in:
Bibliographic Details
Published in:AIP conference proceedings 2022-10, Vol.2661 (1)
Main Authors: Arinicheva, I. V., Arinichev, I. V., Foshchan, G. I.
Format: Article
Language:English
Subjects:
Artificial neural networks
Cereals
Computer vision
Disease
Fungal diseases
Machine learning
Neural networks
Problem solving
Task complexity
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The article is devoted to the main issues of the application of modern mathematical methods for solving problems related to biological objects. An excursion into the history of the emergence of machine learning methods, which today are most in demand for solving the most important tasks of the agro-industrial complex of the Krasnodar Territory, for example, the recognition of dangerous and harmful diseases of cereals at the early stages, is carried out. The possibility of detecting some fungal diseases of rice from photography using machine learning is being investigated. Two diseases are considered: blast disease and a group of diseases - brown spot. To determine the presence of a particular disease in the image, modern computer vision methods based on convolutional neural networks are used. A comparison is made of the four most successful and compact architectural convolutional neural networks: GoogleNet, ResNet-18, SqueezeNet-1.0 and DenseNet-121. It is shown that in the dataset used for the analysis, the disease can be detected with an accuracy of at least 95%.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0107492