Uncertain spatial data handling: Modeling, indexing and query

Managing and manipulating uncertainty in spatial databases are important problems for various practical applications of geographic information systems. Unlike the traditional fuzzy approaches in relational databases, in this paper a probability-based method to model and index uncertain spatial data is proposed. In this scheme, each object is represented by a probability density function (PDF) and a general measure is proposed for measuring similarity between the objects. To index objects, an optimized Gaussian mixture hierarchy (OGMH) is designed to support both certain/uncertain data and certain/uncertain queries. An uncertain R-tree is designed with two query filtering schemes, UR1 and UR2, for the special case when the query is certain. By performing a comprehensive comparison among OGMH, UR1, UR2 and a standard R-tree on US Census Bureau TIGER/Line® Southern California landmark point dataset, it is found that UR1 is the best for certain queries. As an example of uncertain query support OGMH is applied to the Mojave Desert endangered species protection real dataset. It is found that OGMH provides more selective, efficient and flexible search than the results provided by the existing trial and error approach for endangered species habitat search. Details of the experiments are given and discussed.