Hand-draw sketching for image retrieval through fuzzy clustering techniques

Nowadays, the growing of digital media such as images represents an important issue for niultimedia mining applications. Since the traditional information retrieval techniques developed for textual documents do not support adequately these media, new approaches for indexing and retrieval of images are needed. In this paper, we propose an approach for retrieving image by hand-drawn object sketch. For this purpose. we address the classification of images based on shape recognition. The classification is based on the combined use of geometrical and moments features extracted by a given collection of images and achieves shape-based classification through fuzzy clustering techniques. Then, the retrieval is obtained using a hand-draw shape that becomes a query to submit to the system and get ranked similar images.