Improving spatial intersect joins using Symbolic Intersect Detection

Abstract

We introduce a novel technique to drastically reduce the computation required by the refinement step during spatial intersect join processing. This technique, called Symbolic Intersect Detection (SID), detects most of the true hits during a spatial intersect join by scrutinizing symbolic topological relationships between candidate polygon pairs. SID boosts performance by detecting true hits early during the refinement step, thus avoiding expensive polygon intersect computations that would otherwise be required to detect the true hits. Our experimental evaluation with real GIS map data demonstrates that SID can identify more than 80% of the true hits with only minimal overhead. Consequently, SID outperforms known techniques for resolving polygon intersection during the refinement step by more than 50%. Most state-of-the-art methods in spatial join processing can benefit from SID's performance gains because the SID approach integrates easily into the established two-phase spatial join process.KeywordsLine SegmentRefinement StepFilter StepMinimum Bounding RectangleSweep ProcessThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.