Image Converting into Complex Networks : Scale- Level Segmentation Approach

DOI

10.2991/itsmssm-17.2017.88How to use a DOI?

Keywords

images , complex networks, conversion, topology, segmentation , spatial scales

Abstract

Image analysis and recognition is being a contemporary domain for successful tries to apply complex networks as an instrument for thorough studies. Researchers noted that an image having traditionally converted into a network (i.e. taking into account Euclidean distance between pixels only) possesses nodes with similar number of admissible links and the concomitant graph demonstrates a regular topology. As a rule, pixel intensity difference is considered to escape regularity and reach complex property of the network. Contrary revealing more specific traits of an image current study proposes scale segmentation views -local, medium and global - for an image to build a genuine complex network. Case study with two sample images manifests how the scales are connected with formation of a network topology.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Andrey Trufanov
AU  - Nikolay Kinash
AU  - Olga Berestneva
AU  - Alexei Tikhomirov
AU  - Alessandra Rossodivita
PY  - 2017/12
DA  - 2017/12
TI  - Image Converting into Complex Networks : Scale- Level Segmentation Approach
BT  - Proceedings of the IV International research conference "Information technologies in Science, Management, Social sphere and Medicine" (ITSMSSM 2017)
PB  - Atlantis Press
SP  - 417
EP  - 422
SN  - 2352-538X
UR  - https://doi.org/10.2991/itsmssm-17.2017.88
DO  - 10.2991/itsmssm-17.2017.88
ID  - Trufanov2017/12
ER  -