Journal of Siberian Federal University. Engineering & Technologies / Machine Learning Methods for Solving Fire Detection Problems

Full text (.pdf)
Issue
Journal of Siberian Federal University. Engineering & Technologies. 2026 19 (1)
Authors
Kopytov, Artem P.; Kuzin, Denis A.
Contact information
Kopytov, Artem P.: Far Eastern Federal University Vladivostok, Russian Federation; ; Kuzin, Denis A. : Far Eastern Federal University Russian Federation, Vladivostok
Keywords
logistic regression; random forest; convolutional neural network; satellite images; wildfire; spectral channels; error matrix; ROC-AUC metric
Abstract

This article presents an analysis of the machine learning algorithms «Random Forest», «Logistic Regression» and «Convolutional Neural Network». A review of scientific and technical literature on the topic of the study is conducted, the key advantages and limitations of the considered algorithms are analyzed. Particular attention is paid to the application of algorithms in image classification problems, data collection and preparation of satellite images for training. The article contains a practical implementation and comparative analysis of the efficiency of the algorithms in various conditions. The purpose of the article is to provide an understanding of the mathematical foundations of the algorithms and the practical aspects of applying machine learning algorithms to detect fires in satellite images.

Pages
126–138
EDN
WFCOEO
Paper at repository of SibFU
https://elib.sfu-kras.ru/handle/2311/158144

Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).