DESIGN OF AN ONTOLOGY-BASED LOW LEVEL VIDEO ANNOTATION AND RETRIEVAL SYSTEM

Abstract

Advances in technology, as well as the availability of more digital cameras, camcorders and even mobile phones for multimedia content generation, have increased the amount of multimedia data available. However, the systems that currently exist to organise and retrieve these videos are not sufficient to deal with the large and rapidly growing volumes of video. In these thesis, Resource Description Framework(RDF)is used to model the ontology-based system and SPARQL Protocol an RDF Query Language is used to retrieve and manipulate data stored in RDF format.The proposed video retrieval system uses the low level features and the semantic features of a video file as metadata description for video file. In order to further enhance query result, a metadata pool called keywords pool is added to the application. Inverted index structure is used as the text retrieval model to process user queries and video similarity is computed using cosine similarity measure. Experimental results obtained through questionnaires distributed to different levels of system users show a high level of accuracy and effectiveness of the system.The application can be used by individuals, organizations and institutions.