Issue No. 05 - September/October (2002 vol. 14)
<p><b>Abstract</b>—We introduce <it>ImageMap</it>, as a method for indexing and similarity searching in Image DataBases (IDBs). ImageMap answers “queries by example,” involving any number of objects or regions and taking into account their interrelationships. We adopt the most general image content representation, that is, <it>Attributed Relational Graphs</it> (ARGs), in conjunction with the well-accepted <it>ARG editing distance</it> on ARGs. We tested ImageMap on real and realistic medical images. Our method not only provides for visualization of the data set, clustering and data mining, but it also achieves up to 1,000-fold speed-up in search over sequential scanning, with zero or very few false dismissals.</p>
Image database, similarity retrieval, attributed relational graph, editing distance, image indexing.
Euripides G.M. Petrakis, Christos Faloutsos, King-Ip (David) Lin, "ImageMap: An Image Indexing Method Based on Spatial Similarity", IEEE Transactions on Knowledge & Data Engineering, vol. 14, no. , pp. 979-987, September/October 2002, doi:10.1109/TKDE.2002.1033768