Enhancing sports image search and retrieval using multi-modality ontology
Hatem, Yomna; Rasha Ismail; Rady, Sherine; Bahnasy, Khaled;
Abstract
Extracting knowledge from multimedia contents represents recently a big challenge. Organizing and analyzing multimedia collections requires specific tools for extracting knowledge from the contents to enable effective and efficient filtering, searching and retrieval. The use of knowledge models, such as Ontology, is gaining interest among multimedia retrieval researches. This paper builds and integrates a multi-modality ontology to the conventional image annotation and retrieval methodology. The proposed knowledge-model integration highly improves the searching process. Two ontologies are proposed, domain and visual description ontologies. Experiments have demonstrated the efficiency of the proposed multi-modality ontology method when compared against the classical retrieval technique. The results show that using ontologies increases the performance to reach 1, 0.91 and 0.94 for Precision, Recall and F-measure respectively.
Other data
Title | Enhancing sports image search and retrieval using multi-modality ontology | Authors | Hatem, Yomna; Rasha Ismail ; Rady, Sherine ; Bahnasy, Khaled | Keywords | image annotation;semantic searching;multimodality ontology;image retrieval | Issue Date | 28-Jan-2018 | Conference | Proceedings of ICCES 2017 12th International Conference on Computer Engineering and Systems | ISBN | 9781538611913 | DOI | 10.1109/ICCES.2017.8275328 | Scopus ID | 2-s2.0-85046547383 |
Attached Files
File | Description | Size | Format | Existing users please Login |
---|---|---|---|---|
ICCES.2017_Yomna-et-al.pdf | 3 MB | Adobe PDF | Request a copy |
Similar Items from Core Recommender Database
Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.