Отрывок: Thus, we obtained one of the most important advantages of convolutional neural networks - an analog of the convolution operation [14,15]. After that, a study was conducted, the purpose of which was to identify the optimal number of neurons. The criterion of optimality was the criterion of Akaike (AIC) [16]. This criterion fines network for excessive number of param...
Полная запись метаданных
Поле DC Значение Язык
dc.contributor.authorPasynkov, M.K.-
dc.contributor.authorKhachay, M.Y.-
dc.date.accessioned2018-05-15 13:18:44-
dc.date.available2018-05-15 13:18:44-
dc.date.issued2018-
dc.identifierDspace\SGAU\20180514\69217ru
dc.identifier.citationPasynkov M.K. Segmentation of fingerprint images using the simplest neural networks / M.K. Pasynkov and M.Y. Khachay // Сборник трудов IV международной конференции и молодежной школы «Информационные технологии и нанотехнологии» (ИТНТ-2018) - Самара: Новая техника, 2018. - С.1267-1274ru
dc.identifier.urihttp://repo.ssau.ru/handle/Informacionnye-tehnologii-i-nanotehnologii/Segmentation-of-fingerprint-images-using-the-simplest-neural-networks-69217-
dc.descriptionОсновная статьяru
dc.description.abstractSegmentation of fingerprint images is one of the most important problems in automated fingerprint identification system (AFIS). Segmentation is used to separate the area of the fingerprint (foreground) from the background and areas that cannot be recovered. We propose a new algorithm for fingerprint segmentation based on simplest neural networks, binary region labeling technique, and morphological image processing. This approach was tested on public fingerprint dataset provided by Fingerprint Verification Competition (FVC) 2002. The experimental results showed an impressive accuracy was obtained: FAR 2.4%, FRR 1.1% with per-pixel comparison with the reference.ru
dc.language.isoenru
dc.publisherНовая техникаru
dc.subjectfingerprint images, segmentation algorithm, neural networds.ru
dc.titleSegmentation of fingerprint images using the simplest neural networksru
dc.typeArticleru
dc.textpartThus, we obtained one of the most important advantages of convolutional neural networks - an analog of the convolution operation [14,15]. After that, a study was conducted, the purpose of which was to identify the optimal number of neurons. The criterion of optimality was the criterion of Akaike (AIC) [16]. This criterion fines network for excessive number of param...-
Располагается в коллекциях: Информационные технологии и нанотехнологии

Файлы этого ресурса:
Файл Описание Размер Формат  
paper_166.pdfОсновная статья1.98 MBAdobe PDFПросмотреть/Открыть
Показать базовое описание ресурса Просмотр статистики
Поделиться:



Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.