NEURAL NETWORKS FOR SHAPE RECOGNITION BY MEDIAL REPRESENTATION

The article is dedicated to the development of neural networks that process data of a special kind – a medial representation of the shape, which is considered as a special case of an undirected graph. Methods for solving problems that complicate the processing of data of this type by traditional neu...

Full description

Saved in:
Bibliographic Details
Published inInternational archives of the photogrammetry, remote sensing and spatial information sciences. Vol. XLII-2/W12; pp. 137 - 142
Main Authors Lomov, N, Arseev, S
Format Journal Article Conference Proceeding
LanguageEnglish
Published Gottingen Copernicus GmbH 01.01.2019
Copernicus Publications
Subjects
Archives & records
Artificial neural networks
Character recognition
Compression
Data compression
Deep learning
Fonts
Handwriting recognition
Heterogeneity
Neural networks
Representations
Shape
Shape recognition
Online AccessGet full text

Cover

More Information
Summary:The article is dedicated to the development of neural networks that process data of a special kind – a medial representation of the shape, which is considered as a special case of an undirected graph. Methods for solving problems that complicate the processing of data of this type by traditional neural networks – different length of input data, heterogeneity of its structure, unordered constituent elements – are proposed. Skeletal counterparts of standard operations used in convolutional neural networks are formulated. Experiments on character recognition for various fonts, on classification of handwritten digits and data compression using the autoencoder-style architecture are carried out.
ISSN:2194-9034
1682-1750
2194-9034
DOI:10.5194/isprs-archives-XLII-2-W12-137-2019