INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES

Image classification is a classic machine learning task. Deep neural networks are widely used in the field of object classification. However, the problem of analyzing objects with dynamically changing features remains relevant. To solve this problem, the authors propose using a long short-term memor...

Full description

Saved in:
Bibliographic Details
Main Authors: Nikita V. Laptev, Olga M. Gerget, Vladislav V. Laptev, Dmitriy Yu. Kolpashchikov
Format: Article
Language:English
Published: Tomsk Polytechnic University 2023-03-01
Series:Известия Томского политехнического университета: Промышленная кибернетика
Subjects:
Online Access:https://indcyb.ru/journal/article/view/13/12
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849472780945326080
author Nikita V. Laptev
Olga M. Gerget
Vladislav V. Laptev
Dmitriy Yu. Kolpashchikov
author_facet Nikita V. Laptev
Olga M. Gerget
Vladislav V. Laptev
Dmitriy Yu. Kolpashchikov
author_sort Nikita V. Laptev
collection DOAJ
description Image classification is a classic machine learning task. Deep neural networks are widely used in the field of object classification. However, the problem of analyzing objects with dynamically changing features remains relevant. To solve this problem, the authors propose using a long short-term memory networks. Unlike classical convolutional neural networks, the proposed network uses information about the sequence of images, thereby providing a higher classification accuracy of detected objects with dynamic features. In the study, the authors analyze the classification accuracy of smoke cloud detection in a forest using various machine learning methods.
format Article
id doaj-art-768d3d18ce724db8a14558bb1e05f21f
institution Kabale University
issn 2949-5407
language English
publishDate 2023-03-01
publisher Tomsk Polytechnic University
record_format Article
series Известия Томского политехнического университета: Промышленная кибернетика
spelling doaj-art-768d3d18ce724db8a14558bb1e05f21f2025-08-20T03:24:26ZengTomsk Polytechnic UniversityИзвестия Томского политехнического университета: Промышленная кибернетика2949-54072023-03-0111444910.18799/29495407/2023/1/13INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURESNikita V. Laptev0Olga M. Gerget1Vladislav V. Laptev2Dmitriy Yu. Kolpashchikov3National Research Tomsk Polytechnic University, RussiaInstitute of Control Sciences of Russian Academy of Sciences, RussiaNational Research Tomsk Polytechnic University, RussiaNational Research Tomsk Polytechnic University, RussiaImage classification is a classic machine learning task. Deep neural networks are widely used in the field of object classification. However, the problem of analyzing objects with dynamically changing features remains relevant. To solve this problem, the authors propose using a long short-term memory networks. Unlike classical convolutional neural networks, the proposed network uses information about the sequence of images, thereby providing a higher classification accuracy of detected objects with dynamic features. In the study, the authors analyze the classification accuracy of smoke cloud detection in a forest using various machine learning methods.https://indcyb.ru/journal/article/view/13/12neural networkstraditional machine learningclassificationimagedetection of fire hazards
spellingShingle Nikita V. Laptev
Olga M. Gerget
Vladislav V. Laptev
Dmitriy Yu. Kolpashchikov
INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
Известия Томского политехнического университета: Промышленная кибернетика
neural networks
traditional machine learning
classification
image
detection of fire hazards
title INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
title_full INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
title_fullStr INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
title_full_unstemmed INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
title_short INVESTIGATION OF THE CAPABILITIES OF ARTIFICIAL NEURAL NETWORKS WHEN CLASSIFYING OBJECTS DYNAMIC FEATURES
title_sort investigation of the capabilities of artificial neural networks when classifying objects dynamic features
topic neural networks
traditional machine learning
classification
image
detection of fire hazards
url https://indcyb.ru/journal/article/view/13/12
work_keys_str_mv AT nikitavlaptev investigationofthecapabilitiesofartificialneuralnetworkswhenclassifyingobjectsdynamicfeatures
AT olgamgerget investigationofthecapabilitiesofartificialneuralnetworkswhenclassifyingobjectsdynamicfeatures
AT vladislavvlaptev investigationofthecapabilitiesofartificialneuralnetworkswhenclassifyingobjectsdynamicfeatures
AT dmitriyyukolpashchikov investigationofthecapabilitiesofartificialneuralnetworkswhenclassifyingobjectsdynamicfeatures