Systems and methods for automating conversion of drawings to indoor maps and plans

1. A computer-implemented method of determining a geo-location, the method comprising:

determining, by a device, a floor-level outline of a floor of a first building depicted in a CAD drawing;

receiving, by a device, an approximate geo-location of the first building including the floor depicted in the CAD drawing;

obtaining, by a device, an overhead image of a target area encompassing the approximate geo-location of the first building, the overhead image comprising a plurality of buildings within the target area;

identifying, by a device, a plurality of building footprints each associated with a respective building of the plurality of buildings within the target area;

calculating, by a device, a plurality of different types of distance functions that relate the floor-level outline to each of the plurality of building footprints, wherein the calculating creates a plurality of similarity scores corresponding to relationships between the floor-level outline and the plurality of building footprints, and wherein each of the plurality of similarity scores corresponds to a combination of results of the plurality of different types of distance functions;

selecting, by a device, a building footprint from the plurality of building footprints based on the plurality of similarity scores; and

calculating, by a device, a final geo-location of the first building corresponding to the selected building footprint such that the final geo-location of the first building corresponds to a geo-location of one of the respective buildings of the plurality of buildings associated with the selected building footprint.

2. The computer-implemented method of claim 1 wherein calculating the plurality of different types of distance functions further comprises calculating at least two of: a Hausdorf distance; a Modified Hausdorff distance; a Procrustes Shape Analysis; a Frechet distance; and a turning distance.

3. The computer-implemented method of claim 1 wherein calculating the plurality of different types of distance functions further comprises, for each building footprint of the plurality of building footprints, combining a plurality of values indicative of similarity based on a respective plurality of weight values, and wherein the combining results in a similarity score.

4. The computer-implemented method of claim 1 wherein selecting the building footprint further comprises:

determining that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other; and

selecting, from the at least two of the plurality of building footprints, the building footprint closest to the approximate geo-location.

5. The computer-implemented method of claim 4 wherein determining that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other further comprises determining that at least two of the plurality of building footprints have identical similarity scores.

6. The computer-implemented method of claim 1 further comprising correcting the selected building footprint for at least one of: obliqueness of the overhead image; scale of the building footprint in relation to the floor-level outline; and rotational orientation of the building footprint in relation to the floor-level outline.

7. The computer-implemented method of claim 1 wherein determining the floor-level outline further comprises:

preprocessing, by a device, an original CAD drawing and thereby creating a modified CAD drawing, a text database containing text from the modified CAD drawing, a CAD vector-image of the modified CAD drawing, and a CAD raster-image of the modified CAD drawing; and

applying the CAD raster-image, the CAD vector-image, and the text database to a floor-level machine-learning algorithm.

8. The computer-implemented method of claim 1 wherein determining the floor-level outline of the floor depicted in the CAD drawing further comprises receiving the floor-level outline from a user.

9. A computer system comprising:

a processor; and

a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the processor to:

determine a floor-level outline of a floor of a first building depicted in a CAD drawing;

receive an approximate geo-location of the first building including the floor depicted in the CAD drawing;

obtain an overhead image of a target area encompassing the approximate geo-location of the first building, the overhead image comprising a plurality of buildings within the target area;

identify a plurality of building footprints each associated with a respective building of the plurality of buildings within the target area;

calculate a plurality of different types of distance functions that relate the floor-level outline to each of the plurality of building footprints, wherein the calculating creates a plurality of similarity scores corresponding to relationships between the floor-level outline and the plurality of building footprints, and wherein each of the plurality of similarity scores corresponds to a combination of results of the plurality of different types of distance functions;

select a building footprint from the plurality of building footprints, based on the plurality of similarity scores; and

calculate a final geo-location of the first building corresponding to the selected building footprint such that the final geo-location of the first building corresponds to a geo-location of one of the respective buildings of the plurality of buildings associated with the selected building footprint.

10. The computer system of claim 9 wherein when the processor calculates the plurality of different types of distance functions, the instructions cause the processor to calculate at least two of: a Hausdorf distance; a Modified Hausdorff distance; a Procrustes Shape Analysis; a Frechet distance; and a turning distance.

11. The computer system of claim 9 wherein when the processor calculates the plurality of different types of distance functions, the instructions further cause the processor to, for each building footprint of the plurality of building footprints, combine a plurality of values indicative of similarity based on a respective plurality of weight values, and wherein the combining results in a similarity score.

12. The computer system of claim 9 wherein when the processor selects the building footprint, the instructions cause the processor to:

determine that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other; and

select, from the at least two of the plurality of building footprints, the building footprint closest to the approximate geo-location.

13. The computer system of claim 12 wherein when the processor determines that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other, the instructions further cause the processor to determine that at least two of the plurality of building footprints have identical similarity scores.

14. The computer system of claim 9 wherein the instructions further cause the processor to correct the selected building footprint for at least one of: obliqueness of the overhead image; scale of the building footprint in relation to the floor-level outline; and rotational orientation of the building footprint in relation to the floor-level outline.

15. The computer system of claim 9 wherein when the processor determines the floor-level outline, the instructions cause the processor to:

preprocess an original CAD drawing and thereby creating a modified CAD drawing, a text database containing text from the modified CAD drawing, a CAD vector-image of the modified CAD drawing, and a CAD raster-image of the modified CAD drawing; and

apply the CAD raster-image, the CAD vector-image, and the text database to a floor-level machine-learning algorithm.

16. The computer system of claim 9 wherein when the processor determines the floor-level outline of the floor depicted in the CAD drawing, the instructions cause the processor to receive the floor-level outline from a user.

17. A computer-implemented method of determining a geo-location, the method comprising:

determining, by a device, a floor-level outline of a floor of a first building depicted in a CAD drawing;

receiving, by a device, an approximate geo-location of the first building including the floor depicted in the CAD drawing;

obtaining, by a device, an overhead image of a target area encompassing the approximate geo-location of the first building, the overhead image comprising a plurality of buildings within the target area;

identifying, by a device, a plurality of building footprints each associated with a respective building of the plurality of buildings within the target area;

calculating, by a device, a plurality of different types of distance functions that relate the floor-level outline to each of the plurality of building footprints, wherein the calculating results in a plurality of values indicative of similarity, and wherein the calculating uses and combines at least two distance functions selected from a group comprising a Hausdorf distance, a Modified Hausdorff distance, a Procrustes Shape Analysis, a Frechet distance, and a turning distance;

for each building footprint of the plurality of building footprints, combining the plurality of values indicative of similarity based on a respective plurality of weight values, wherein the combining results in a plurality of similarity scores corresponding to relationships between the floor-level outline and the plurality of building footprints;

selecting, by a device, a building footprint from the plurality of building footprints, based on the plurality of similarity scores; and

calculating, by a device, a final geo-location of the first building corresponding to the selected building footprint such that the final geo-location of the first building corresponds to a geo-location of one of the respective buildings of the plurality of buildings associated with the selected building footprint.

18. The computer-implemented method of claim 17 wherein selecting the building footprint further comprises:

determining that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other; and

selecting, from the at least two of the plurality of building footprints, the building footprint closest to the approximate geo-location.

19. The computer-implemented method of claim 18 wherein determining that at least two of the plurality of building footprints have similarity scores within a predetermined range of each other further comprises determining that at least two of the plurality of building footprints have identical similarity scores.

20. The computer-implemented method of claim 17 further comprising correcting the selected building footprint for at least one of: obliqueness of the overhead image; scale of the building footprint in relation to the floor-level outline; and rotational orientation of the building footprint in relation to the floor-level outline.