Indexing and searching in multimedia database management system

Searching a digital library having large volumes of digital images, texts, and graphics or video sequences has become important in this visual age. To make the searching of such data objects, efficient and effective indexing and reverse indexing techniques are very important. These techniques are also required for proper query optimization. In the proposed system the indexing and reverse indexing technique of multimedia database has been highlighted to find the entire nearest neighbor points in a given multimedia data sets. This paper introduces a new index structure RMD-Tree that answers both NNMD (nearest neighbor multimedia data set) and RNNMD (reverse nearest neighbor multimedia data set) queries in multimedia database. Here, a single index structure is employed for a dynamic database, in contrast to the use of multiple indexes. This enables significant savings of time in dynamically maintaining the index structure. The proposed index structure outperforms a significant margin (more than 90% in terms of number of leaf nodes accessed) in reverse nearest neighbor queries. It also shows improvement in NNMD queries over standard techniques. Furthermore, performance in insertion and deletion is significantly enhanced by the ability to combine multiple queries (NNMD&RNNMD) in one traversal of the tree. These facts make the proposed index structure extremely preferable in both static and dynamic cases.