Enhanced spatial index for point in polygon operations
View Patent ↗An improved method is provided for determining whether a sample point is within a defined geographic area. Indexes for the geographic area of interest are generated in advance. Such indexes complement the traditional spatial indexing techniques such as quad tree and r-tree. The geographic area, as defined by an outer boundary, is subdivided into some regular geometric shape, preferably a rectangle, encoded into a suitable form, and indexed. Then, a simplified comparison of the sample point to the indexed regular shapes is made.
1. A computer implemented method for determining whether a sample point is within a geographic area by generating indices for subdivided areas within the geographic area, and the geographic area having a defined boundary, the method comprising:
defining a rectangular boundary box that encloses the geographic area;
dividing the boundary box into a plurality of equal first level rectangular boxes at a first dividing level;
determining for each of the first level boxes whether the first level box is (a) entirely outside the geographic area; (b) entirely inside the geographic area; or (c) intersecting with the defined boundary;
for each of the first level boxes performing the following indexing steps:
if the first level box is (a) entirely outside the geographic area, then that box is discarded and it is not part of the indices;
if the first level box is (b) entirely inside the geographic area, then that box is added to the indices and is assigned an index number that is stored with the corresponding range of coordinates for that box;
if the first level box is (c) intersecting with the defined boundary then dividing that box into a plurality of equal second level rectangular boxes at the second dividing level;
repeating the determining and indexing steps for each of the second level boxes as described above for the first level boxes;
receiving the sample point for which a response is required as to whether the sample point is located within the geographic area;
comparing the sample point to the coordinate ranges of the indexed boxes to determine if the sample point is included in the indices;
if the sample point is within the indexed boxes then providing a response indicating that the sample point is within the geographic area; and
if the sample point is not within the indexed boxes then performing a full geometric analysis of the sample point in comparison to a geometry of the defined boundary.
2. The method of claim 1 wherein the step of repeating further comprises:
iteratively repeating the determining and indexing steps for a subsequent number of dividing levels until (1) there are no more boxes that intersect with the defined boundary, or (2) a maximum dividing level has been reached.
3. The method of claim 2 wherein the indexing step includes including the level of an indexed box as part of the index number.
4. The method of claim 1 further including, prior to the step of comparing the sample point to the coordinate ranges of the Indexed boxes, determining whether the sample point is inside or outside the boundary box; and
if the sample point is outside the boundary box, then providing an output indicating that the sample point is outside of the geographic area;
if the sample point is inside the boundary box, then proceeding with the comparing step and subsequent steps thereafter.
5. The method of claim 1 wherein conventional analysis includes comparing the sample point to a complete geometry of the defined boundary.
6. The method of claim 1 wherein multiple indices are generated for multiple layers of maps having different boundaries for a same physical location, and wherein the step of comparing the sample point to the coordinate ranges of the Indexed boxes is done concurrently for the multiple Indices.
7. The method of claim 6 wherein the multiple layers represent different types of insurance risks, and whereby the method determines whether the sample point is in located in designated risk zones in each of the respective layers.
8. A computer implemented method for determining whether a sample point is within a geographic area by generating indices for subdivided areas within the geographic area, and the geographic area having a defined boundary, the method comprising:
defining a geometric boundary shape that encloses the geographic area;
dividing the boundary shape into a plurality of equal first level shapes at a first dividing level;
determining for each of the first level shapes whether the first level shape is (a) entirely outside the geographic area; (b) entirely inside the geographic area; or (c) intersecting with the defined boundary;
for each of the first level shapes performing the following indexing steps:
if the first level shape is (a) entirely outside the geographic area, then that shape is discarded and it is not part of the indices;
if the first level shape is (b) entirely inside the geographic area, then that shape is added to the indices and is assigned an index number that is stored with the corresponding range of coordinates for that shape;
if the first level shape is (c) intersecting with the defined boundary then dividing that shape into a plurality of equal second level shapes at the second dividing level;
repeating the determining and indexing steps for each of the second level shapes as described above for the first level shapes;
receiving the sample point for which a response is required as to whether the sample point is located within the geographic area;
comparing the sample point to the coordinate ranges of the indexed shapes to determine if the sample point is included in the indices;
if the sample point is within the indexed shapes then providing a response indicating that the sample point Is within the geographic area; and
if the sample point is not within the indexed shapes then performing a full geometric analysis of the sample point in comparison to a geometry of the defined boundary.
9. The method of claim 8 wherein the step of repeating further comprises:
iteratively repeating the determining and indexing steps for a subsequent number of dividing levels until (1) there are no more shapes that Intersect with the defined boundary, or (2) a maximum dividing level has been reached.
10. The method of claim 9 wherein the indexing step includes including the level of an indexed shape as part of the index number.
11. The method of claim 8 further including, prior to the step of comparing the sample point to the coordinate ranges of the indexed shapes, determining whether the sample point is inside or outside the boundary shape; and
if the sample point is outside the boundary shape; then providing an output indicating that the sample point is outside of the geographic area;
if the sample point is inside the boundary shape, then proceeding with the comparing step and subsequent steps thereafter.
12. The method of claim 8 wherein conventional analysis includes comparing the sample point to a complete geometry of the defined boundary.
13. The method of claim 8 wherein multiple Indices are generated for multiple layers of maps having different boundaries for a same physical location, and wherein the step of comparing the sample point to the coordinate ranges of the indexed shapes is done concurrently for the multiple indices.
14. The method of claim 13 wherein the multiple layers represent different types of insurance risks, and whereby the method determines whether the sample point is in located in designated risk zones in each of the respective layers.