Near-duplicate image recognition

Near-duplicate image recognition

Bibliografski detalji
Parent link:Mechanical Engineering, Automation and Control Systems (MEACS).— 2014.— [4 p.]
Glavni autor: Nemirovskiy V. B. Viktor Borisovich
Autor kompanije: Национальный исследовательский Томский политехнический университет (ТПУ) Институт кибернетики (ИК) Кафедра информатики и проектирования систем (ИПС)
Daljnji autori: Stoyanov A. K. Aleksandr Kirillovich
Sažetak:Title screen
Usage of multi-step segmentation for near-duplicate image recognition is investigated. The search pattern based on the rank distributions of the brightness clusters cardinality is suggested. Experimental results on the near-duplicate image recognition based on the application of the suggested search pattern are given. It is shown that the use of a multi-step segmentation and rank distributions of the brightness clusters cardinality allows to successfully recognize the duplicates, that received by the considerable visual distortion of the original image.
Режим доступа: по договору с организацией-держателем ресурса
Jezik:engleski
Izdano: 2014
Teme:
электронный ресурс
труды учёных ТПУ
image
pixel
point map
recurrent neural network
clustering
segmentation
image recognition
near-duplicates
rank distribution
пиксели
нейронные сети
кластеризация
сегментация
распознавание образов
изображения
Online pristup:http://dx.doi.org/10.1109/MEACS.2014.6986916
Format: Elektronički Poglavlje knjige
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=639059

Internet

http://dx.doi.org/10.1109/MEACS.2014.6986916

Slični predmeti