Публікація:
An effective method for transforming an image description into a compact vector for classification

Спрощена сторінка документа
dc.contributor.authorGorokhovatskyi, V.
dc.contributor.authorTvoroshenko, I.
dc.date.accessioned2024-12-20T18:41:55Z
dc.date.available2024-12-20T18:41:55Z
dc.date.issued2024
dc.description.abstractThe work develops method to solve a fundamental problem in computer vision: image recognition of visual objects. Based on the implementation of the distance matrix model, it was possible to form effective integrated features in the form of one-dimensional data distributions and vectors for the sum of the matrix columns, which reduced computational costs without losing the effectiveness of classification on the training data sample. The efficiency of image classification was experimentally evaluated using software modeling.
dc.description.sponsorshipThe results of this research were obtained under the international research project INITIATE under the grant No. 101136775-HORIZON-WIDERA-2023-ACCESS-03.
dc.identifier.citationGorokhovatskyi V. An effective method for transforming an image description into a compact vector for classification / V. Gorokhovatskyi, I. Tvoroshenko // Information Technology and Implementation (Satellite) : Conference Proceedings, November 21, 2024. – Kyiv : Publishing House «Caravela», 2024. - pp. 25-28.
dc.identifier.isbn978-960-801-870-9
dc.identifier.urihttps://openarchive.nure.ua/handle/document/29476
dc.language.isoen_US
dc.publisherPublishing House «Caravela»
dc.subjectcomputer vision
dc.subjectdescriptor
dc.subjectdistance matrix
dc.subjectkeypoint
dc.subjectorthogonal decomposition
dc.subjectprocessing speed
dc.subjectrelevance of descriptions
dc.titleAn effective method for transforming an image description into a compact vector for classification
dc.typeConference proceedings
dspace.entity.typePublication

Файли

Оригінальний пакет
Зараз показано 1 - 1 з 1
Немає доступних мініатюр
Назва:
tvoroshenko.pdf
Розмір:
72.05 KB
Формат:
Adobe Portable Document Format
Завантажити
Ліцензійний пакет
Зараз показано 1 - 1 з 1
Немає доступних мініатюр
Назва:
license.txt
Розмір:
9.55 KB
Формат:
Item-specific license agreed upon to submission
Опис:
Завантажити

Колекції

Кафедра інформатики (ІНФ)