Guided geometry extraction 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;

obtaining feature geometries of a fingerprint from a fingerprint database, the fingerprint being in proximity to the location in the path network;

orienting, using a processor of the end-user device, the depth map with the feature geometries of the fingerprint;

comparing control points of an oriented feature geometry from the depth map with control points within the fingerprint; and

calculating an individual match rate for the oriented feature geometry based on an average deviation in distance between each control point of the oriented feature geometry from the depth map and respective control points in the fingerprint.

2. The method of claim 1 , wherein the proximity is initially determined by global positioning system (GPS) data or Wi-Fi positioning data.

3. The method of claim 1 , wherein the control points within the fingerprint are identified for comparison by defining a bounding box encompassing latitude, longitude, and altitude coordinates of each respective control point in the oriented feature geometry from the depth map.

4. The method of claim 1 , wherein the depth map is oriented by (1) shifting the depth map along a road segment in the path network, both parallel and perpendicular to a direction of travel to align the depth map with a collection location for the fingerprint database, and (2) rotating the depth map to align the depth map with a capture direction at the collection location.

5. The method of claim 1 , further comprising:

saving the selected feature geometry from the fingerprint database for further analysis when the average deviation is lower than a predefined threshold value, therein providing an acceptable match rate.

6. The method of claim 5 , further comprising:

comparing additional control points from additional feature geometries within the depth map with respective control points within the fingerprint;

calculating additional individual match rates for each additional feature geometry; and

identifying certain additional feature geometries with acceptable match rates until a predefined number of feature geometries have been identified.

7. The method of claim 6 , wherein at least one additional feature geometry is identified from a separate quadrant surrounding the device.

8. The method of claim 7 , further comprising

calculating an overall match rate based on each individual match rate; and

determining a geographic location of the end-user device in the path network through a multilateration calculation when the overall match rate exceeds a predefined threshold minimum match rate value.

9. The method of claim 8 , further comprising:

collecting, by the end-user device, a new depth map;

receiving new feature geometries in a new fingerprint from the fingerprint database, the new fingerprint being selected based upon the determined geographic location.

10. The method of claim 9 , wherein the collecting of the new depth map by the end-user device is computed to be aligned at a same geographic location in the road network as a depth map collected for the fingerprint database using the determined geographic location.

11. The method of claim 1 , wherein the oriented feature geometry is a highest ranking feature geometry within a defined zone surrounding the end-user device, wherein ranking values for each extracted feature geometry are based on one or more of the following factors: (1) shape of the feature geometries, (2) size of the feature geometries, (3) elevation of the feature geometries, or (4) variance of the feature geometries.

12. A method comprising:

receiving a depth map at a location in a path network from an end-user device;

retrieving feature geometries of a fingerprint from a fingerprint database, the fingerprint being in proximity to the location in the path network;

orienting, using a processor of the fingerprint database, the depth map with the feature geometries of the fingerprint;

comparing control points of extracted feature geometries from the depth map with control points within the fingerprint; and

calculating an individual match rate for the oriented feature geometry based on an average deviation in distance between each control point of the oriented feature geometry from the depth map and respective control points in the fingerprint.

13. The method of claim 12 , further comprising:

calculating an overall match rate based on each individual match rate;

determining a geographic location of the end-user device in the path network through a multilateration calculation when the overall match rate exceeds a predefined threshold minimum match rate value; and

transmitting the determined geographic location to the end-user device over a connected network.

14. 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 in a path network;

obtain feature geometries of a fingerprint from a fingerprint database, the fingerprint being in proximity to the location in the path network;

orient the depth map with the feature geometries of the fingerprint;

compare control points of an oriented feature geometry from the depth map with control points within the fingerprint; and

calculate an individual match rate for the oriented feature geometry based on an average deviation in distance between each control point of the oriented feature geometry from the depth map and respective control points in the fingerprint.

15. The apparatus of claim 14 , wherein the depth map is oriented by (1) shifting the depth map along a road segment in the path network, both parallel and perpendicular to a direction of travel to align the depth map with a collection location for the fingerprint database, and (2) rotating the depth map to align the depth map with a capture direction at the collection location.

16. The apparatus of claim 14 , wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

saving the selected feature geometry from the fingerprint database for further analysis when the average deviation is lower than a predefined threshold value, therein providing an acceptable match rate.

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

compare additional control points from additional extracted feature geometries within the depth map with respective control points within the fingerprint;

calculate additional individual match rates for each additional extracted feature geometry; and

identify additional feature geometries with acceptable match rates until a predefined number of feature geometries have been identified.

18. The apparatus of claim 17 , further comprising

calculating an overall match rate based on each individual match rate; and

determining a geographic location of the end-user device in the path network through a multilateration calculation when the overall match rate exceeds a predefined threshold minimum match rate value.

19. The apparatus of claim 18 , wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

collect a new depth map;

receive new feature geometries in a new fingerprint from the fingerprint database, the new fingerprint being selected based upon the determined geographic location.

20. The apparatus of claim 19 , wherein the collecting of the new depth map by the end-user device is computed to be aligned at a same geographic location in the road network as a depth map collected for the fingerprint database using the determined geographic location.