Saliency Region Detection via Local and Global Features

DOI

10.2991/csic-15.2015.27How to use a DOI?

Keywords

Salient detection, Local contrast, Global contrast, Weight value

Abstract

Saliency detection plays an important role in many applications such as image segmentation, retrieval, editor and object recognition. Salience detection methods mainly divide into two categories: local contrast method and global contrast method. AC method is one of the typically local contrast methods, and HC method is a classical method base on global contrast. In this paper, we proposed an approach combined with AC and HC method, it not only considers local features but global information. After synthesizing the salience map of the AC and HC methods, firstly, according to the characteristics of human visual system, we pay attention to the salience center of an image, and add a weight value to each pixel, the closer to the center, the higher the weight. Then, we enhance the contrast of the image, receding darker part and increasing bright one. To confirm the proposed approach's effectiveness, we compare AC and HC method with the proposed one. The experimental results show that the new method generates a better salience map.

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  - Dan Hu
AU  - Yan Wang
AU  - Shaohui Qian
AU  - Weiyu Yu
PY  - 2015/07
DA  - 2015/07
TI  - Saliency Region Detection via Local and Global Features
BT  - Proceedings of the 2015 International Conference on Computer Science and Intelligent Communication
PB  - Atlantis Press
SP  - 117
EP  - 120
SN  - 2352-538X
UR  - https://doi.org/10.2991/csic-15.2015.27
DO  - 10.2991/csic-15.2015.27
ID  - Hu2015/07
ER  -