A probability-based uncertainty model for point-in-polygon analysis in GIS

In a geographical information system (GIS), a conventional point-in-polygon analysis is used to determine whether a point is located inside a polygon with a Boolean result. Due to positional uncertainties existing with both the point and the polygon, the Boolean answer cannot describe the relationship of closeness between the point and the polygon. This paper aims to develop a model and to provide a continuous index-the probability value-to indicate the extent of the uncertain point located inside the uncertain polygon based upon existing research development. This probability index is derived based on probability and statistical theories considering the statistical uncertainty distributions of the point and the polygon's vertices. The associated mathematical expressions for the probability index are addressed in cases depending on the intersection between the polygon and the error ellipse of the point. The proposed probability index can provide an objective description of the relationship of closeness between an uncertain point and an uncertain polygon.