(An) efficient filtering method for partial similarity search in image databases

Abstract

We propose the CXHistogram and CXSim() as an image representation and a similarity measure, respectively, for efficient partial similarity search in image databases. By using CXHistogram and CXSim(), XMage and HIPAS are introduced in this thesis as a model for partial similarity searching in image databases. XMage is an image matching method based on the partial similarity, while HIPAS is a method for subimage matching application. Region-based image retrieval is a method of retrieving partially similar images. It has been proposed as a way to accurately process queries in an image database. In region-based image retrieval, region matching is indispensable for computing the partial similarity between two images because the query processing is based upon regions instead of the entire image. A naive method of region matching is a sequential comparison between regions, which causes severe overhead and deteriorates the performance of query processing. In this thesis, a new image contents representation, called CXHistogram(Condensed eXtended Histogram), is presented in conjunction with a well-defined distance function CXSim() on the CXHistogram. The CXSim() is a new image-to-image similarity measure to compute the partial similarity between two images. It achieves the effect of comparing regions of two images by simply comparing the two images. The CXSim() reduces query space by pruning irrelevant images, and it is used as a filtering function before sequential scanning. Extensive experiments were performed on real image data to evaluate XMage and HIPAS. It provides a significant pruning of irrelevant images with no false dismissals. As a consequence, XMage and HIPAS achieve up to 5.9-fold and 4.84-fold speed-up, respectively, in search over the $R^*$-tree search followed by sequential scanning.