Ontology based classification for multi-label image annotation

Reshma, Ismat Ara; Ullah, Md Zia; Aono, Masaki

Abstract

Image annotation has been an important task for visual information retrieval. It usually involves a multi-class multi-label classification problem. To solve this problem, many researches have been conducted during last two decades, although most of the proposed methods rely on the training data with the ground truth. To prepare such a ground truth is an expensive and laborious task that cannot be easily scaled, and “semantic gaps” between low-level visual features and high-level semantics still remain. In this paper, we propose a novel approach, ontology based supervised learning for multi-label image annotation, where classifiers' training is conducted using easily gathered Web data. Moreover, it takes advantage of both low-level visual features and high-level semantic information of given images. Experimental results using 0.507 million Web images database show effectiveness of the proposed framework over existing method.

Citation

Reshma, I. A., Ullah, M. Z., & Aono, M. (2014). Ontology based classification for multi-label image annotation. In 2014 international conference of advanced informatics: concept, theory and application (ICAICTA) (226-231). https://doi.org/10.1109/ICAICTA.2014.7005945

• Conference Name: 2014 International Conference of Advanced Informatics: Concept, Theory and Application (ICAICTA)
• Conference Location: Bandung, Indonesia
• Start Date: Aug 20, 2014
• End Date: Aug 21, 2014
• Online Publication Date: Jan 12, 2015
• Publication Date: 2014
• Deposit Date: Mar 13, 2023
• Publisher: Institute of Electrical and Electronics Engineers
• Pages: 226-231
• Book Title: 2014 international conference of advanced informatics: concept, theory and application (ICAICTA)
• DOI: https://doi.org/10.1109/ICAICTA.2014.7005945
• Keywords: noisy training data, classification, image annotation, ontology