Brief Analysis on Typical Image Saliency Detection Methods

DOI

10.2991/icismme-15.2015.336How to use a DOI?

Keywords

saliency detection; computer vision; spatial domain; frequency domain;image segmentation.

Abstract

Image saliency detection is becoming more and more important in the field of computer vision. The purpose is to detect salient objects in complex background, at the same time, reduce time complexity to get salient objects with high-resolution and clear edge. In this paper some brief analysis on the state-of-the-art studies relating to image saliency detection has been implemented at first. Next, two kinds of the relating methods has been separated, the former is based on space domain and the latter is based on frequency domain. And then some typical methods respective in space domain and frequency domain are introduced and tested in image databases used for image saliency detection, the experimental results of these method are analyzed with running time and saliency map and the results of image segmentation using fixed threshold and adaptive threshold are compared. Finally, the conclusion and prospects of studies on image saliency detection are proposed.

Copyright

© 2015, 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  - Wenwen Pan
AU  - Xiaofei Sun
AU  - Xia Wang
AU  - Wei Zhang
PY  - 2015/07
DA  - 2015/07
TI  - Brief Analysis on Typical Image Saliency Detection Methods
BT  - Proceedings of the First International Conference on Information Sciences, Machinery, Materials and Energy
PB  - Atlantis Press
SP  - 1612
EP  - 1617
SN  - 1951-6851
UR  - https://doi.org/10.2991/icismme-15.2015.336
DO  - 10.2991/icismme-15.2015.336
ID  - Pan2015/07
ER  -