Use of a convolutional neural network to auto-determine a floor height and floor height elevation of a building

1. A computer-implemented method of detecting a first floor height (FFH) of a first floor of a subject building, the method comprising:

extracting, with a computer, digital surface model (DSM) information from a database, the DSM information includes surface elevation information of the subject building on a parcel of land on which the subject building is located;

applying an overhead image of the subject building to a CNN-based AI engine implemented on the computer, which has previously been trained, so as to identify a first floor of the subject building from the image, the CNN-based AI engine having previously been trained with other images of a plurality of other buildings, the other images including overhead images of the plurality of other buildings, the CNN-based AI engine including a data extraction network having a first feature extracting layer, a region-of-interest (ROI) pooling layer, and a data vectorizing layer;

analyzing the image with the CNN-based AI engine to estimate the FFH of the subject building, the analyzing including

applying by the first feature extracting layer a first convolutional operation to the image to generate a subject feature map,

pooling features of the subject feature map to create a ROI-pooled feature map, the features including a roof of the subject building detected in the overhead image,

identifying a roof elevation of the roof of the subject building from the DSM information, and

determining an interior height differential between the roof elevation and the first floor of the subject building; and

estimating the FFH of the subject building as a difference between the roof elevation and the height differential between the roof elevation and first floor, the estimating includes applying a first fully connected operation to the pooling regions of the ROI-pooled feature map.

2. The method of claim 1 , wherein the other images previously applied to the CNN-based AI engine include interior images of interior spaces of the plurality of other buildings.

3. The method of claim 2 , wherein the analyzing includes the CNN-based AI engine identifying regions of interest (ROIs) in the interior images of the interior spaces of the plurality of other buildings.

4. The method of claim 3 , wherein the ROIs include at least one of an appliance, a countertop, or a piece of furniture.

5. The method of claim 3 , wherein the ROIs includes an interior structure.

6. The method of claim 5 , wherein the interior structure includes at least one of an interior staircase, an entrance, a door, or a doorstep.

7. The method of claim 1 , further comprising:

extracting previously stored data from another database that includes interior property information for the subject building.

8. The method of claim 7 , wherein the another database is a multiple listing service (MLS) database.

9. The method of claim 7 , wherein the interior property information includes at least one of room dimensional measurements, number of floors, foundation type, property type, basement information, structure design drawing, or a 3D design model.

10. The method of claim 1 , further comprising:

training the CNN-based AI engine by inputting other images as training images and ground truth images to the CNN-based AI engine, and backpropagating losses so as to establish data extraction parameters for a data extraction network portion of the CNN-based AI engine.

11. A method of detecting a first floor height (FFH) of a first floor of a subject building, the method comprising:

applying an image of the subject building to a CNN-based AI engine, implemented on a computer, which has previously been trained, so as to identify a first floor of the subject building from the image, the CNN-based AI engine having previously been trained with other images of a plurality of other buildings, the other images including at least one of a front, a side, a roof, or a back-side view of individual buildings of the plurality of other buildings, the CNN-based AI engine including a data extraction network having a first feature extracting layer, a region-of-interest (ROI) pooling layer, and a data vectorizing layer, the analyzing including;

analyzing the image with the CNN-based AI engine and determining the FFH of the subject building;

applying by the first feature extracting layer a first convolutional operation to the image to generate a subject feature map,

pooling features of the subject feature map, with the ROI pooling layer to create a ROI-pooled feature map, the features including the front of the subject building detected in the image, and

estimating the FFH of the subject building as a difference between a roof elevation and the height differential between the roof elevation and first floor, the estimating includes applying a first fully connected operation to the pooling regions of the ROI-pooled feature map;

extracting, with the computer, digital surface model (DSM) information from a database, the DSM information includes surface elevation information of the terrain and/or surface elevation information from a dataset for the parcel of land on which the subject building is located;

converting the FFH of the subject building to a first floor elevation (FFE) from the FFH and the DSM information; and

identifying a height of the subject building and the FFE of the subject building based on the height of the subject building.

12. The method of claim 11 , further comprising:

obtaining information on a building footprint of the subject building; and

identifying a location at an adjacent grade point along the building footprint so as to determine an elevation of the location and the FFE of the subject building at the location.

13. The method of claim 12 , wherein the obtaining includes determining a geometry of the building footprint from a perimeter of a roof line of the subject building.

14. The method of claim 11 , further comprising:

training the CNN-based AI engine, the training including inputting other images as training images and ground truth images to the CNN-based AI engine, and backpropagating losses so as to establish data extraction parameters for a data extraction network portion of the CNN-based AI engine.

15. The method of claim 14 , wherein the training further comprises:

training the CNN-based AI engine to detect building types including applying training images of different types of buildings to the CNN-based AI engine, the different types of buildings including basement-below grade, basement-daylight, and basement-walkout.

16. The method of claim 14 , wherein the training further comprises:

training the CNN-based AI engine to detect foundation types by applying training images of buildings with different types of foundations.

17. The method of claim 14 , wherein the training of the CNN-based AI engine further includes training the CNN-based AI engine with overhead images of buildings having roofs and walkways leading to the main entrance.

18. The method of claim 14 , wherein the training further comprises:

training the CNN-based AI engine to estimate a height of a main entrance of the subject building, and the image being a street view image of the subject building, wherein the image includes an object detected in the image as at least one of steps and a ramp leading to the main entrance, and the height of the main entrance of the subject building is estimated from an estimated height of the objected detected in the image.

19. The method of claim 11 , further comprising:

extracting previously stored data in another database that includes interior property information for the subject building.

20. The method of claim 19 , wherein the interior property information includes at least one of room dimensional measurements, number of floors, foundation type, property type, basement information, structure design drawing, or a 3D design model.