GRF: Greedy Based Relevance Feedback Algorithm for Retrieval of Multimedia Object

The perspective of multimedia object retrieval is the matching of relevant object in the image collection based on user query in a large database. In the past years, there is a comprehensive enhancement under content-based image retrieval system (CBIR).However, performance of the system need to be faster. Recent work has described various approaches and schemes that would have been expensive and difficult to arrange. There are several other methods which include query expression according to user needs. From the catalogue of retrieval techniques, Relevance Feedback enriches query refinement process. This paper proposes an algorithm for relevance feedback based on greedy approach to enhance the retrieval method. The greedy algorithm is used to sum up the performance of multimedia object retrieval system by using Relevance feedback technique. The main focus of the paper is to analyze how this greedy aspect of Relevance Feedback can be consolidated into existing retrieval system.