A mobile multimedia database system for infants education environment

Abstract

To effectively deal with video data, a semantic-based retrieval scheme that allows for processing diverse user queries and saving them on the database is required. This paper proposes a semantic-based multimedia database system that enables users to search the meaning of video data in a diverse manner. It uses both semantic and dependency weights to perform video retrieval for the environment education of infants. In the proposed system, the user searches multimedia data for environment education through entering keywords. The mobile agent then computes both semantic and dependency weights, ensuring the accuracy of data retrieved using calculated weights as annotative information of key frames. As a result of implementing and testing the prototype of the proposed system, a higher precision of approximately 96.5% was obtained. © Springer-Verlag Berlin Heidelberg 2005.