Bitmap indexing method for complex similarity queries with relevance feedback

The similarity indexing and searching is well known to be a difficult one for high-dimensional applications such as multimedia databases. Especially, it becomes more difficult when multiple features have to be indexed together. Moreover, few indexing methods are currently available to effectively support disjunctive queries for relevance feedback.In this paper, we propose a novel indexing method that is designed to efficiently handle complex similarity queries as well as relevance feedback in high-dimensional image and video databases. In order to provide the indexing method with the flexibility in control multiple features and multiple query objects, our method treats every dimension independently. The efficiency of our method is realized by a specialized bitmap indexing that represents all objects in a database as a set of bitmaps. The percentage of data accessed in our indexing method is inversely proportional to the overall dimensionality, and thus the performance deterioration with the increasing dimensionality does not occur.Our main contributions are three-fold: (1) We provide a novel way to index high-dimensional data; (2) Our method efficiently handles complex similarity queries; and (3) Disjunctive queries driven by relevance feedback are efficiently treated. Our empirical results demonstrate that our indexing method achieves speedups of 10 to 15 over the linear scan.