Summarizing Spatial Relations - A Hybrid Histogram

Summarizing topological relations is fundamental to many spatial applications including spatial query optimization. In this paper, we examine the selectivity estimation for range window query to summarize the four important topological relations: contains, contained, overlap, and disjoint. We propose a novel hybrid histogrammethod which uses the concept of Min-skew partition in conjunction with Euler histogram approach. It can effectively model object spatial distribution. Our extensive experiments against both synthetic and real world datasets demonstrated that our hybrid histogram techniques improve the accuracy of the existing techniques by about one order of magnitude while retaining the cost efficiency.