Granted Patent B2

US 12,738,052 · App. 18/793,500 · Granted Sep 15, 2026

Inventors: Shantanu Rajora (San Francisco, CA); Nitin Das (Gurugram, IN)

Assignee: AIDash Inc

G06V20/188G06T7/0002G06T7/62G06T17/05G06V10/26G06V10/764G06V10/82G06V20/13G06V20/176G06T2207/10032G06T2207/20081G06T2207/20084G06T2207/30184G06T2207/30188

Abstract

Claims (78)

1 . A non-transitory computer-readable medium comprising executable instructions, the executable instructions being executable by one or more processors to perform a method, the method comprising:

receiving multiple images for a geographic area, the geographic area including multiple electrical assets of a power distribution infrastructure;

identifying the multiple electrical assets in the multiple images;

identifying multiple trees in the multiple images;

determining a digital terrain model by performing minimum pooling operations on a digital surface model until a peak signal to noise ratio of the digital terrain model to the digital surface model does not exceed a threshold;

generating a canopy height model based on the digital surface model and the digital terrain model for the geographic area;

determining, based on the canopy height model for the geographic area, a height of a particular tree of the multiple trees;

determining a distance between the particular tree and a particular electrical asset of the multiple electrical assets;

identifying, based on the height and the distance, the particular tree as a potential hazard; and

providing a notification of the particular tree as the potential hazard.

2 . The non-transitory computer-readable medium of claim 1 , wherein the method further comprises:

receiving a set of georeferenced stereo satellite images of the geographic area; and

generating the digital surface model of the geographic area based on the set of georeferenced stereo satellite images of the geographic area.

3 . The non-transitory computer-readable medium of claim 1 , wherein the method further comprises generating or modifying a vegetation trim plan for a portion of the geographic area that includes the particular tree.

4 . The non-transitory computer-readable medium of claim 1 , wherein the method further comprises modifying a frequency of occurrence of an existing vegetation trim plan for a portion of the geographic area that includes the particular tree.

5 . The non-transitory computer-readable medium of claim 1 wherein generating the canopy height model based on the digital surface model and the digital terrain model for the geographic area includes:

generating an initial canopy height model by subtracting the digital terrain model from the digital surface model;

determining first pixels in the initial canopy height model that have heights greater than a second threshold;

determining second pixels in the digital terrain model corresponding to the first pixels and replacing heights of the second pixels with null values;

determining third pixels surrounding the second pixels in the digital terrain model;

generating a final digital terrain model by interpolating heights from the third pixels and replacing the heights of the second pixels with interpolated heights; and

generating the canopy height model by subtracting the final digital terrain model from the digital surface model.

6 . The non-transitory computer-readable medium of claim 1 , wherein the method further comprises:

generating one or more segmentation maps by providing the multiple images to one or more fully convolutional networks to classify at least some pixels of the multiple images as either trees or non-trees;

generating one or more rasters of the geographic area based on the one or more segmentation maps and the multiple images;

generating one or more vectors based on the one or more rasters, a vector including one or more polygons, a polygon representing a tree and having a set of coordinates defining the polygon;

determining, based on the canopy height model, heights of the multiple trees in the geographic area; and

associating the heights of the multiple trees with the one or more polygons of the one or more vectors.

7 . The non-transitory computer-readable medium of claim 1 , wherein the method further comprises determining a triangle defined by the height, a distance from a center of mass of the particular tree to the particular electrical asset, and a calculated hypotenuse, and wherein identifying, based on the height and the distance, the particular tree as the potential hazard includes identifying, based on the triangle, the particular tree as the potential hazard.

8 . The non-transitory computer-readable medium of claim 1 wherein determining, based on the canopy height model for the geographic area, the height of the particular tree includes determining, based on the canopy height model for the geographic area, multiple heights of the particular tree, and wherein the height is a maximum height of the particular tree.

9 . The non-transitory computer-readable medium of claim 1 , the method further comprising:

determining an area of the particular tree; and

determining a volume of the particular tree using the height and the area.

10 . The non-transitory computer-readable medium of claim 9 , the method further comprising identifying a species of the particular tree, and wherein determining the volume of the particular tree using the height and the area includes determining the volume of the particular tree using the height, the area, and the species.

11 . The non-transitory computer-readable medium of claim 1 wherein generating the canopy height model based on the digital surface model and the digital terrain model for the geographic area includes subtracting the digital terrain model from the digital surface model to generate the canopy height model for the geographic area.

12 . A method, comprising:

receiving multiple images for a geographic area, the geographic area including multiple electrical assets of a power distribution infrastructure;

identifying the multiple electrical assets in the multiple images;

identifying multiple trees in the multiple images;

determining a digital terrain model by performing minimum pooling operations on a digital surface model until a peak signal to noise ratio of the digital terrain model to the digital surface model does not exceed a threshold;

generating a canopy height model based on the digital surface model and the digital terrain model for the geographic area;

determining, based on the canopy height model for the geographic area, a height of a particular tree of the multiple trees;

determining a distance between the particular tree and a particular electrical asset of the multiple electrical assets;

identifying, based on the height and the distance, the particular tree as a potential hazard; and

providing a notification of the particular tree as the potential hazard.

13 . The method of claim 12 , further comprising:

receiving a set of georeferenced stereo satellite images of the geographic area; and

generating the digital surface model of the geographic area based on the set of georeferenced stereo satellite images of the geographic area.

14 . The method of claim 12 , further comprising generating or modifying a vegetation trim plan for a portion of the geographic area that includes the particular tree.

15 . The method of claim 12 , further comprising modifying a frequency of occurrence of an existing vegetation trim plan for a portion of the geographic area that includes the particular tree.

16 . The method of claim 12 wherein generating the canopy height model based on the digital surface model and the digital terrain model for the geographic area includes:

generating an initial canopy height model by subtracting the digital terrain model from the digital surface model;

determining first pixels in the initial canopy height model that have heights greater than a second threshold;

determining second pixels in the digital terrain model corresponding to the first pixels and replacing heights of the second pixels with null values;

determining third pixels surrounding the second pixels in the digital terrain model;

generating a final digital terrain model by interpolating heights from the third pixels and replacing the heights of the second pixels with interpolated heights; and

generating the canopy height model by subtracting the final digital terrain model from the digital surface model.

17 . The method of claim 12 , further comprising:

generating one or more segmentation maps by providing the multiple images to one or more fully convolutional networks to classify at least some pixels of the multiple images as either trees or non-trees;

generating one or more rasters of the geographic area based on the one or more segmentation maps and the multiple images;

generating one or more vectors based on the one or more rasters, a vector including one or more polygons, a polygon representing a tree and having a set of coordinates defining the polygon;

determining, based on the canopy height model, heights of the multiple trees in the geographic area; and

associating the heights of the multiple trees with the one or more polygons of the one or more vectors.

18 . The method of claim 12 , further comprising determining a triangle defined by the height, a distance from a center of mass of the particular tree to the particular electrical asset, and a calculated hypotenuse, and wherein identifying, based on the height and the distance, the particular tree as the potential hazard includes identifying, based on the triangle, the particular tree as the potential hazard.

19 . The method of claim 12 , further comprising determining, based on the canopy height model for the geographic area, the height of the particular tree includes determining, based on the canopy height model for the geographic area, multiple heights of the particular tree, and wherein the height is a maximum height of the particular tree.

20 . The method of claim 12 , further comprising:

determining an area of the particular tree; and

determining a volume of the particular tree using the height and the area.

21 . A system comprising at least one processor and memory containing executable instructions, the executable instructions being executable by the at least one processor to:

receive multiple images for a geographic area, the geographic area including multiple electrical assets of a power distribution infrastructure;

identify the multiple electrical assets in the multiple images;

identify multiple trees in the multiple images;

determine a digital terrain model by performing minimum pooling operations on a digital surface model until a peak signal to noise ratio of the digital terrain model to the digital surface model does not exceed a threshold;

generate a canopy height model based on the digital surface model and the digital terrain model for the geographic area;

determine, based on the canopy height model for the geographic area, a height of a particular tree of the multiple trees;

determine a distance between the particular tree and a particular electrical asset of the multiple electrical assets;

identify, based on the height and the distance, the particular tree as a potential hazard; and

provide a notification of the particular tree as the potential hazard.

Assignments (1)

Continuity (4)

References Cited (48)