Experimental Studies on the Recognition of Small-Sized Objects in Video Images Using Multidimensional Spatial-Subband Vectors
DOI:
https://doi.org/10.52575/2687-0932-2022-49-2-432-440Keywords:
decision function, evaluation, recognition, experimental studies, vector, covariance matrix, subbandAbstract
A decisive rule has been developed for recognizing small-sized objects in video images, which allows recognizing various small-sized objects in video images with high quality indicators. The input data for the decisive rule are samples of spatially subband vectors formed from the image of objects. Experimental studies of the decisive function are carried out using images with various small-sized objects located on them. The obtained numerical values of the logarithm of the likelihood ratio used to make a decision on object recognition are given. Experimental studies have shown that the largest values of the logarithm of the likelihood ratio are located proportionally to those pixels of the image under study on which the object on which the training was conducted is located. The developed decisive rule makes it possible to recognize various small-sized objects on video images with high quality indicators. The developed approach to the construction of the decision rule allows us to use optimal solutions and use the Neumann-Pearson criterion to set the threshold level. Experimental studies using in-situ data confirm the capabilities of the developed decisive rule for the recognition of small-sized objects in video images.
Downloads
References
Васин Ю.Г., Лебедев Л.И. 2007. Распознавание объектов составных изображений на основе структурных и корреляционно-экстремальных методов. Сборник докладов 13-й Всероссийской конференции «Математические методы распознавания образов» (MMRO-13), Зеленогорск, 148: 285–288.
Гиренко А.В., Ляшенко В.В., Машталир В.П., Путятин Е.П. 1996. Методы корреляционного обнаружения объектов. Харьков, АО «Бизнесинформ»: 112.
Фомин Я.А., Тарловский Г.Р. 1986. Статистическая теория распознавания образов. Москва, Радио и связь: 264.
Фурман Я.А. 2004. Введение в контурный анализ и его применение к обработке изображений и сигналов, ФИЗМАТЛИТ: 456.
Burdanova E.V., Zhilyakov E.G., Mamatov A.V., Nemtsev A.N., Oleynik I.I. 2019. Decisive rule experimental studies to detect objects on the background of the earth surface using polarization differences of radar signals. COMPUSOFT. An International Journal of Advanced Computer Technology, 8(6): 3166–3170.
Goloschapova V.A., Kalashnikov P.A., Oleynik I.I. 2020. Parametric decision rules for object recognition in multi-dimensional vector representation of color images. Science, Educatition, Practice: materials of the International University Science Forum (Canada, Toronto), May 27: 238–246.
Zalivin A.N., Oleynik I.I., Pirogenko Y.A. 2020. Decision rule for recognizing small objects based on subband processing of radar signals. Science, Educatition, Practice: materials of the International University Science Forum (Canada, Toronto), April 22: 197–207.
Zhilyakov E.G. 2015. Optimal subband methods for analyzing and synthesizing signals of finite duration. Automation and Telemechanics, 4: 51–66.
Zhilyakov E.G., Belov S.P., Oleinik I.I., Babarinov S.L., Trubitsyna D.I. 2020. Generalized sub band analysis and signal synthesis. Bulletin of Electrical Engineering and Informatics, 1(9): 78–86.
Zhilyakov E.G., Belov S.P., Oleinik I.I., Prokhorenko E.I. 2019. Regularization of Inverse Signal Recovery Problems, HELIX the Scientific Explorer, 9 (2): 4883–4889.
Abstract views: 97
Share
Published
How to Cite
Issue
Section
Copyright (c) 2022 ECONOMICS. INFORMATION TECHNOLOGIES
This work is licensed under a Creative Commons Attribution 4.0 International License.
