Selecting feature geometries for localization of a device

1. A method comprising:

collecting, by an end-user device, a depth map at a location in a path network;

extracting, using a processor of the end-user device, two-dimensional feature geometries from the depth map;

ranking the extracted feature geometries to provide ranking values for the extracted feature geometries;

selecting a portion of the extracted feature geometries based upon the ranking values and a geographic distribution of the extracted feature geometries;

transmitting the selected portion of the extracted feature geometries to an external processor; and

receiving, from the external processor, a geographic location of the end-user device through a comparison of the selected portion of the extracted feature geometries and a database of feature geometries for the path network.

2. The method of claim 1 , wherein the ranking values for the extracted feature geometries are based upon one or more of the following properties: shapes, sizes, elevations from a road level, variance, or colors.

3. The method of claim 2 , wherein the ranking values are based on the shapes of the feature geometries, wherein a ranking value is higher for a shape identified as a line, arc, or spline in comparison with a shape not identified as a line, arc, or spline.

4. The method of claim 2 , wherein the ranking values are based on the elevations of the feature geometries, wherein a ranking value is higher for a feature geometry at a higher elevation in comparison with a feature geometry at a lower elevation.

5. The method of claim 2 , wherein the ranking values are based on the sizes of the feature geometries, wherein a ranking value is higher for a larger sized feature geometry in comparison with a lower sized feature geometry.

6. The method of claim 2 , wherein the ranking values are based on the variance of the feature geometries, wherein a ranking value is higher for an invariant feature geometry in comparison with a variant feature geometry.

7. The method of claim 1 , wherein the geographic distribution of the extracted feature geometries comprises a number of zones of the depth map of equal volume surrounding the end-user device, and the selecting comprises an identification of at least one feature geometry from each zone.

8. The method of claim 7 , wherein the number of zones is at least 4.

9. The method of claim 1 , wherein the geographic distribution of the extracted feature geometries comprises a number of zones of the depth map of equal volume surrounding the end-user device, and

wherein the selecting comprises an identification of an extracted feature geometry from each zone having a highest ranking value, wherein the ranking value of each extracted feature geometry is based upon one or more of the following properties: a shape of the feature geometry, a size of the feature geometry, an elevation from a road level of the feature geometry, a variance of the feature geometry, or a color of the feature geometry.

10. The method of claim 1 , wherein the ranking values for the extracted feature geometries are based upon a weighted combination of (1) a shape of each extracted feature geometry, (2) a size of each extracted feature geometries, (3) an elevation from a road level of each extracted feature geometry, and (4) a variance of each extracted feature geometry.

11. The method of claim 10 , wherein a higher weight is provided for each of the elevation and the size in comparison with each of the shape and the variance.

12. A method comprising:

receiving a depth map of a location in a path network;

extracting, using a processor, two-dimensional feature geometries from the depth map; and

ranking the extracted feature geometries to provide ranking values for the extracted feature geometries at the location;

sorting at least a portion of the extracted feature geometries based upon the ranking values of the extracted feature geometries from highest ranking value to lowest ranking value; and

encoding the sorted feature geometries into a fingerprint database for the location in the path network.

13. The method of claim 12 , wherein the sorting the portion of the extracted feature geometries is additionally based upon a geographic distribution of the extracted feature geometries.

14. The method of claim 12 , wherein the ranking values for the extracted feature geometries are based upon one or more of the following properties: shapes, sizes, elevations from a road level, variance, or colors.

15. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

collect a depth map at a location of an end-user device in a path network;

extract two-dimensional feature geometries from the depth map;

rank the extracted feature geometries to provide ranking values for the extracted feature geometries;

select a portion of the extracted feature geometries based upon the ranking values and a geographic distribution of the extracted feature geometries; and

transmit the selected portion of the extracted feature geometries to an external processor.

16. The apparatus of claim 15 , wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

receive, from the external processor, a geographic location of the end-user device through a comparison of the selected portion of the extracted feature geometries and a database of feature geometries for the path network.

17. The apparatus of claim 15 , wherein the ranking values for the extracted feature geometries are based upon one or more of the following properties: a shape of the feature geometry, a size of the feature geometry, an elevation from a road level of the feature geometry, a variance of the feature geometry, or a color of the feature geometry.

18. The apparatus of claim 15 , wherein the ranking values for the extracted feature geometries are based upon a weighted combination of (1) a shape of each extracted feature geometry, (2) a size of each extracted feature geometries, (3) an elevation from a road level of each extracted feature geometry, and (4) a variance of each extracted feature geometry.

19. The apparatus of claim 18 , wherein a higher weight is provided for each of the elevation and the size in comparison with each of the shape and the variance.

20. The apparatus of claim 15 , wherein the geographic distribution of the extracted feature geometries comprises a number of zones of equal volume surrounding the end-user device, and the selecting comprises an identification of at least one feature geometry from each zone.