Table | Card | RUSMARC | |
Allowed Actions: –
Action 'Read' will be available if you login or access site from another network
Action 'Download' will be available if you login or access site from another network
Group: Anonymous Network: Internet |
Annotation
В данной работе описаны основные понятия и методы распознавания образов. Изучены основные виды искусственных нейронных сетей для задач распознавания цифровых образов, проведен анализ. Рассмотрен фреймворк Hadoop его файловая система HDFS, потока данных MapReduce и библиотека deeplearning4j. Спроектирована приложение для распознавания образов с помощью фреймворка Hadoop и наборов данных MNIST. Разработаны классы для приложения с работой в параллельном режиме. Приложение протестировано на данных из интернета.
In the given work the basic concepts and methods of image recognition are described. The main types of artificial neural networks for the tasks of recognizing images are studied, the analysis is carried out. The Hadoop framework, its file system HDFS, analysis of the MapReduce data flow and the library Deeplearning4j are investigated. An application for image recognition was designed using the Hadoop framework and datasets. Classes have been developed for the application working in parallel mode. The application has been tested on random images from the Internet.
Document access rights
Network | User group | Action | ||||
---|---|---|---|---|---|---|
ILC SPbPU Local Network | All |
![]() ![]() ![]() |
||||
External organizations N2 | All |
![]() |
||||
External organizations N1 | All | |||||
Internet | Authorized users SPbPU |
![]() ![]() ![]() |
||||
Internet | Authorized users (not from SPbPU, N2) |
![]() |
||||
Internet | Authorized users (not from SPbPU, N1) | |||||
![]() |
Internet | Anonymous |
Usage statistics
|
Access count: 2
Last 30 days: 0 Detailed usage statistics |