A fuzzy approach to complex linguistic query based image retrieval

Most content based techniques in vogue retrieve images based on an example image or features extracted from images. Retrieval based on linguistic queries has not received much attention. We present a fuzzy connective approach to handle complex linguistic queries consisting of multiple attributes. We represent each attribute in a complex query by a (multi dimensional) membership function. The degree to which an image satisfies the attribute is given by the membership value of the feature vector corresponding to the image in the membership function for the attribute. We propose the use of fuzzy connectives to combine the query to arrive at an overall degree of satisfaction to rank images for retrieval. We report experimental results on a database of 2384 texture images created from the VisTex texture database available at the MIT media lab. Our results indicate that this approach can be used to facilitate meaningful image retrieval based on complex linguistic queries