Image Retrieval Relevance Feedback algorithms: Trends and Techniques
P. Ganesh Kumar · 2013
Abstract: With many applications, Content based Image Retrieval (CBIR) has come into the attention in recent decades. To reduce the schematic gap a wide variety of relevance feedback (RF) algorithms have been developed in recent years to improve the performance of CBIR systems. These RF algorithms capture user’s preferences and bridge the semantic gap. Many schemes and techniques of relevance feedback exist with many assumptions and operating criteria. Yet there exist few ways of quantitatively measuring and comparing different relevance feedback algorithms. Such analysis is necessary if a CBIR system is to perform consistently. In this paper, different RF techniques are reviewed. The selection of papers include sources from image processing journals, conferences, books, dissertations and thesis out of more than 500 journals, books and online research databases. The state of art research on each category is provided with emphasis on developed technologies and image properties used by them. Finally, conclusions are drawn summarizing commonly used techniques and their complexities in applications. The visual features of the old lady image and the dog image are very similar, but their semantic meanings are totally different as shown in fig 1. Another kind of example for semantic gap is shown in fig 2.For the left query image of fig 2, some users focus on the sea beach so the best match could be a sea beach like the one in the middle image; while others may focus on the coconut tree, so the rightmost image is the best match. These problems are come under semantic gap. Fig 1. Examples of semantic gap I.