A Fuzzy System for Content Based Retrieval

Content based retrieval in multimedia databases has recently gained much interest. Image data is the most commonly used multimedia data type. Various content based retrieval systems for image data have been developed based on image features such as texture, color, edge, and shape that can be extracted from the image. Three major inherent characteristics of the existing content based retrieval systems are imprecision, a large number of feature values and the user's preferences. Existing content based retrieval systems deal with "imprecision" providing partial information through user interfaces such as "retrieve all the images that have color similar to this". Moreover, these systems, developed on visual properties, lack of formal models. Moreover, these systems deal with all possible values of visual properties such as color. The visual properties are extracted from the image and stored into the database as an indexable summary information. The users are unable to express their needs i..