The enhancements in multimedia technology have accelerated the development of new techniques for querying and retrieval in video databases. Based on these new techniques, effective user interfaces are designed especially for query specification. Since the Internet is a suitable multiple client environment, most of the video database systems are designed to work over the Internet.In this thesis, a web-based user interface for a video database system is presented. The interface is composed of a hierarchical windows-based structure where logically separable sub-queries are handled in separate windows. For example, spatial and motion queries are handled separately. The separation of the interface aims to facilitate the query specification process. Furthermore, the user is allowed to get partial results for the sub-queries. The retrieval and result presentation process displays the query results in the form of video frame sequences on a separate window with the capability of playing each result separately. As another contribution of this thesis, considering the fact that most of the systems include querying by object features, a novel approach for color and shape querying and retrieval is proposed. The proposed approach is histogram-based and it is shown through the performance experiments that it overcomes the deficiencies of the existing methods. This new approach is supported by an auxiliary object extraction tool in the query-by-feature sub-system of the video database system. The object extraction tool is semi-automatic, hence it can successfully capture salient object regions for most of the images and/or video frames.

Most image database prototypes and products focus mainly on similarity searches over syntactic features of images. DISIMA aims at providing querying on both syntactic and semantic features of images. The content of an image is viewed as a set of salient objects (regions of interest). Salient objects are organized into two levels: physical salient objects that store syntactic features and logical salient objects that give the semantics. DISIMA integrates a declarative query language (MOQL) and a visual query language VisualMOQL. 1.

This paper presents a multimedia join operator that is carried out through the method of the nearest neighbor search. In contrast to related approaches that utilizes a similarity function to perform a join between two instances of the input tables, we adopt the more flexible and widely used nearest neighbor method. First, we introduce a simple nearest neighbor search algorithm based on an nested-loop execution strategy, second an optimized version is proposed which takes advantage of query point clustering in a hypersphere. Several experiments are performed to demonstrate the e#ciency of the optimized algorithm over the simple one for different datasets, datasizes and dimensions. Key words: Multimedia Databases, Processing a Multimedia Joins, Nearest Neighbor Search 1