Tree Fusion Method for Semantic Concept Detection in Images

Jafar MANSOURI; Morteza KHADEMI

doi:10.1587/transinf.E97.D.2209

Tree Fusion Method for Semantic Concept Detection in Images

Jafar MANSOURI, Morteza KHADEMI

Full Text Views

0

Share
Cite this

Summary :

A novel fusion method for semantic concept detection in images, called tree fusion, is proposed. Various kinds of features are given to different classifiers. Then, according to the importance of features and effectiveness of classifiers, the results of feature-classifier pairs are ranked and fused using C4.5 algorithm. Experimental results conducted on the MSRC and PASCAL VOC 2007 datasets have demonstrated the effectiveness of the proposed method over the traditional fusion methods.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.8 pp.2209-2211

Publication Date: 2014/08/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.2209

Type of Manuscript: LETTER

Category: Image Processing and Video Processing

Authors

Jafar MANSOURI
Ferdowsi University of Mashhad
Morteza KHADEMI
Ferdowsi University of Mashhad

Keyword

semantic concept detection in images, fusion, C4.5 algorithm

Cite this

Copy

Jafar MANSOURI, Morteza KHADEMI, "Tree Fusion Method for Semantic Concept Detection in Images" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 8, pp. 2209-2211, August 2014, doi: 10.1587/transinf.E97.D.2209.
Abstract: A novel fusion method for semantic concept detection in images, called tree fusion, is proposed. Various kinds of features are given to different classifiers. Then, according to the importance of features and effectiveness of classifiers, the results of feature-classifier pairs are ranked and fused using C4.5 algorithm. Experimental results conducted on the MSRC and PASCAL VOC 2007 datasets have demonstrated the effectiveness of the proposed method over the traditional fusion methods.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E97.D.2209/_p

Copy

@ARTICLE{e97-d_8_2209,
author={Jafar MANSOURI, Morteza KHADEMI, },
journal={IEICE TRANSACTIONS on Information},
title={Tree Fusion Method for Semantic Concept Detection in Images},
year={2014},
volume={E97-D},
number={8},
pages={2209-2211},
abstract={A novel fusion method for semantic concept detection in images, called tree fusion, is proposed. Various kinds of features are given to different classifiers. Then, according to the importance of features and effectiveness of classifiers, the results of feature-classifier pairs are ranked and fused using C4.5 algorithm. Experimental results conducted on the MSRC and PASCAL VOC 2007 datasets have demonstrated the effectiveness of the proposed method over the traditional fusion methods.},
keywords={},
doi={10.1587/transinf.E97.D.2209},
ISSN={1745-1361},
month={August},}

Copy

TY - JOUR
TI - Tree Fusion Method for Semantic Concept Detection in Images
T2 - IEICE TRANSACTIONS on Information
SP - 2209
EP - 2211
AU - Jafar MANSOURI
AU - Morteza KHADEMI
PY - 2014
DO - 10.1587/transinf.E97.D.2209
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2014
AB - A novel fusion method for semantic concept detection in images, called tree fusion, is proposed. Various kinds of features are given to different classifiers. Then, according to the importance of features and effectiveness of classifiers, the results of feature-classifier pairs are ranked and fused using C4.5 algorithm. Experimental results conducted on the MSRC and PASCAL VOC 2007 datasets have demonstrated the effectiveness of the proposed method over the traditional fusion methods.
ER -