Interest Point Detection in Video Surveillance Image Based on Topology Structure Features of Directed and Weight Complex Network

DOI

10.2991/emcs-17.2017.134How to use a DOI?

Keywords

Complex network;Image;Degree;Betweenness;Interest point

Abstract

Extracting and describing the key information of images are the focus of the image research. Interest point detection is an important part of image target recognition process. This paper presents a new method for detecting interest points in digital images. In this paper, the pixels in the image is taken as the network node, and the similarity and distance between the pixels are taken as the weight of the edge, and then the image is expressed as the directional and weighted network. We propose a new network node feature evaluation standard based on the integration of degree and betweenness. According to this standard, we extract the key nodes in the network, which are interest points of image. In order to verify the effectiveness of our algorithm, we use the algorithm in the complex video surveillance images. The results show that our algorithm could identify interest nodes from the images effectively.

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  - Qingyu Zou
AU  - Baisheng Liu
AU  - Jianwen Guan
AU  - Yan Chen
PY  - 2017/03
DA  - 2017/03
TI  - Interest Point Detection in Video Surveillance Image Based on Topology Structure Features of Directed and Weight Complex Network
BT  - Proceedings of the 2017 7th International Conference on Education, Management, Computer and Society (EMCS 2017)
PB  - Atlantis Press
SP  - 700
EP  - 704
SN  - 2352-538X
UR  - https://doi.org/10.2991/emcs-17.2017.134
DO  - 10.2991/emcs-17.2017.134
ID  - Zou2017/03
ER  -