A study on image searching method in super distributed database

In this paper we propose a new searching method for image information in a super-distributed environment. First, we show that the performance of the search can be evaluated by entropy reduction, and propose a new searching method based on entropy reduction, called the EBS method. Second, we apply the EBS method to image searching using various kinds of query schemes, for example, keyword search, color-pair matching, and browsing. Third, the design concept and experimental results of the prototype system are introduced.