DMCA
Avrithis - "A Region Thesaurus Approach for High-Level Concept Detection (2007)
| Venue: | in the Natural Disaster Domain", 2nd international conference on Semantics And digital Media Technologies (SAMT |
| Citations: | 9 - 9 self |
BibTeX
@INPROCEEDINGS{Spyrou07avrithis-,
author = {Evaggelos Spyrou and Yannis Avrithis},
title = {Avrithis - "A Region Thesaurus Approach for High-Level Concept Detection},
booktitle = {in the Natural Disaster Domain", 2nd international conference on Semantics And digital Media Technologies (SAMT},
year = {2007}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract. This paper presents an approach on high-level feature de-tection using a region thesaurus. MPEG-7 features are locally extracted from segmented regions and for a large set of images. A hierarchical clustering approach is applied and a relatively small number of region types is selected. This set of region types defines the region thesaurus. Using this thesaurus, low-level features are mapped to high-level con-cepts as model vectors. This representation is then used to train support vector machine-based feature detectors. As a next step, latent semantic analysis is applied on the model vectors, to further improve the anal-ysis performance. High-level concepts detected derive from the natural disaster domain. 1
Keyphrases
high-level concept detection region thesaurus approach region type region thesaurus model vector high-level feature de-tection low-level feature next step natural disaster domain support vector machine-based feature detector hierarchical clustering approach high-level concept large set latent semantic analysis high-level con-cepts mpeg-7 feature anal-ysis performance small number
