Similarity measurement for aggregation of spatial objects
Byungwoo Kim, Jong Pil Yoon · 2005
SVG (Scalable Vector Graphics) is increasingly important to display multimedia data over the Internet. Although SVG is powerful in exchange, transformation, and rendering of image data, it has still the weakness of content-based component-sensitive image search mainly because it does not provide yet the descriptors of (component) image objects and their spatial relationships. Also, although similarity matches have been studied for long, similarity measurements for images that contain component image objects are not successfully developed. In this paper, we describe a new approach of a similarity measurement of aggregated image objects.