Fuzzy logic based interpretation and fusion of color queries

The paper presents a novel fuzzy logic-based approach for interpretation of color queries and a technique for fusion of multiple queries. Currently, in most real world systems, users have to provide the content of color in terms of percentage or many real numbers to retrieve an image. In this paper a fuzzy logic-based interface is presented which allows user to express queries in terms of phrases like mostly red, many green, few red. The paper also presents a novel way of fusing multiple queries. By incorporating the fusion of multiple queries, it was noticed that there was a prominent increase in the confidence factor for the images. Fusion experiments were conducted with neuro-fuzzy, fuzzy AND and binary AND techniques. A comparative analysis showed that neuro-fuzzy has outperformed other two techniques.