Biodiversity image retrieval framework for monogeneans

Abstract

In this paper, we propose a framework for Biodiversity Image Retrieval using Content Based Image Retrieval (CBIR) and structured vocabulary. The monogenean images used in this paper belong to the order Dactylogyridea Bychowsky, 1937. We use illustrations of monogenean haptoral bar diagnostic hard part to demonstrate how textual and content-based information can be integrated in building a better retrieval system. In this approach, an ontology-based image annotation and retrieval is developed to support the CBIR. Using ontology in combination with CBIR reduces the number of images needed in the training set, utilizing the provided user parameters. Efficiency of retrieval of the proposed approach is calculated using the relevance ranking and classification error rate of the retrieved images; and the efficiency of overall retrieval performance.

Key words:

• biodiversity image retrieval
• content-based image retrieval
• image illustrations
• monogenean
• ontology

Acknowledgements

This project was supported by the University of Malaya's Postgraduate Research Fund (PS284/2009B) to the first author and the University of Malaya Research Grant (RG053/09SUS) to the second and fourth authors.

Associate Editor: Elliot Shubert