Skip to Main Content
The present paper describes a query-by-sketch image retrieval system aimed at reducing the semantic gap between low-level features and high-level semantics by adopting relevance feedback. To reduce the semantic gap in this content-based image retrieval system, user sketches play an important role in relevance feedback. When users mark similar images of output images with "relevant" labels, these "relevant" images are relevant to the sketch image in feedback. On the other hand, when users mark dissimilar images of output images with "irrelevant" labels, these "irrelevant" images are irrelevant to the sketch image in feedback. However, in the method the effect of the feedback has been limited by the images with "relevant" and "irrelevant" labels. The purpose of the present paper is to expand the effect on other images, using query-by-sketch retrieval and rough sets. The proposed method is applied to 7,500 images in Corel Photo Gallery. Experimental results show that the proposed method is effective in retrieving images.