Towards Visual Vocabulary and Ontology-based Image Retrieval System
Résumé
Several approaches have been introduced in image retrieval field. However, many limitations, such as the semantic gap, still exist. As our motivation is to improve image retrieval accuracy, this paper presents an image retrieval system based on visual vocabulary and ontology. We propose, for every query image, to build visual vocabulary and ontology based on images annotations. Image retrieval process is performed by integrating both visual and semantic features and similarities.
Origine | Fichiers produits par l'(les) auteur(s) |
---|