DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL

A model of text image detector based on a convolutional neural network architecture is presented, capable of synthesizing high-level features of images in the «black box» mode. An implementation of the detector application, based on algorithms of multi-scale scanning and local responses interpretati...

Full description

Saved in:
Bibliographic Details
Main Author: N. N. Kuzmitsky
Format: Article
Language:Russian
Published: National Academy of Sciences of Belarus, the United Institute of Informatics Problems 2016-09-01
Series:Informatika
Online Access:https://inf.grid.by/jour/article/view/15
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849243303084556288
author N. N. Kuzmitsky
author_facet N. N. Kuzmitsky
author_sort N. N. Kuzmitsky
collection DOAJ
description A model of text image detector based on a convolutional neural network architecture is presented, capable of synthesizing high-level features of images in the «black box» mode. An implementation of the detector application, based on algorithms of multi-scale scanning and local responses interpretation is described, allowing to find out text samples on images of real scenes. Advantages in comparison with analogs are shown and efficiency evaluation on an example of a known database is conducted.
format Article
id doaj-art-38fffacfe92b4b4a9735a6baae70fa77
institution Kabale University
issn 1816-0301
language Russian
publishDate 2016-09-01
publisher National Academy of Sciences of Belarus, the United Institute of Informatics Problems
record_format Article
series Informatika
spelling doaj-art-38fffacfe92b4b4a9735a6baae70fa772025-08-20T03:59:31ZrusNational Academy of Sciences of Belarus, the United Institute of Informatics ProblemsInformatika1816-03012016-09-0102122114DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODELN. N. Kuzmitsky0Брестский государственный технический университетA model of text image detector based on a convolutional neural network architecture is presented, capable of synthesizing high-level features of images in the «black box» mode. An implementation of the detector application, based on algorithms of multi-scale scanning and local responses interpretation is described, allowing to find out text samples on images of real scenes. Advantages in comparison with analogs are shown and efficiency evaluation on an example of a known database is conducted.https://inf.grid.by/jour/article/view/15
spellingShingle N. N. Kuzmitsky
DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
Informatika
title DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
title_full DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
title_fullStr DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
title_full_unstemmed DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
title_short DETECTION OF TEXT OBJECTS IN IMAGES OF REAL SCENES BASED ON CONVOLUTIONAL NEURAL NETWORK MODEL
title_sort detection of text objects in images of real scenes based on convolutional neural network model
url https://inf.grid.by/jour/article/view/15
work_keys_str_mv AT nnkuzmitsky detectionoftextobjectsinimagesofrealscenesbasedonconvolutionalneuralnetworkmodel