Utilizing a 3D scanner to estimate damage to a roof

• US 10,013,720 B1
• Filed: 12/09/2015
• Issued: 07/03/2018
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Original Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Inventors: Plummer, Bryan Allen, Cross, Drew, Tofte, Nathan L.
• Primary Examiner(s): Ebersman, Bruce I
• Application Number:
  US14/964,195
• Time in Patent Office: 937 Days
• Field of Search: 705 4
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06Q 20/10 specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
• CPC Class Codes: G06Q 30/0278 Product appraisal
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G06T 15/08 Volume rendering
• CPC Class Codes: G06T 17/00 Three dimensional [3D] modelling, e.g. data description of 3D objects
• CPC Class Codes: G06T 2210/56 Particle system, point based geometry or rendering
• CPC Class Codes: G06T 2219/012 Dimensioning, tolerancing
• CPC Class Codes: G06T 7/00 Image analysis
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: H04N 13/204 using stereoscopic image cameras stereoscopic photography G03B35/00

Abstract

A damage assessment module operating on a computer system automatically evaluates a roof, estimating damage to the roof by analyzing a point cloud of a roof. The damage assessment module identifies individual shingles from the point cloud and detects potentially damaged areas on each of the shingles. The damage assessment module then maps the potentially damaged areas of each shingle back to the point cloud to determine which areas of the roof are damaged. Based on the estimation, the damage assessment module generates a report on the roof damage.

Claims

1. A system comprising:

(A) one or more 3-dimensional (3D) scanners for generating a 3D point cloud representing a roof;

(B) one or more processors; and

(C) one or more memory devices communicatively connected to the one or more processors, the one or more memories including the 3D point cloud generated via the one or more 3D scanners;

wherein the one or more memory devices further include computer-readable instructions that, when executed, cause the one or more processors to;

(i) retrieve from the one or more memories the 3D point cloud generated via the one or more 3D scanners;

(ii) utilize region growing segmentation to identify from the 3D point cloud a plurality of sets of points, each set of points representing an individual shingle on the roof, including doing the following for each set of points;

identify a group of points from the 3D point cloud;

determine a first measurement of a surface shape of the group of points;

identify an additional one or more points from the 3D point cloud;

add the additional one or more points to the group of points;

determine a second measurement of the surface shape of the group of points; and

remove the additional one or more points from the group of points when the second measurement of the surface shape exceeds a threshold;

wherein the first measurement and second measurement each relate to at least one of;

(i) curvature;

or (ii) smoothness;

(iii) analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle; and

(iv) generate and display a report indicating the damaged shingle is damaged, wherein the displayed report includes one or more of;

(a) a textual representation of damage done to the damaged shingle;

or (b) a graphical representation of damage done to the damaged shingle.

2. The system of claim 1, wherein the instructions to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle comprise instructions to:

compare the set of points to the other sets within the plurality of sets of points to determine that the set of points represents an outlier relative to the other sets.

3. The system of claim 1, wherein the instructions to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle comprise instructions to:

generate from the plurality of sets of points a model 3D point cloud representing a model shingle.

4. The system of claim 3, wherein the instructions to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprise instructions to:

perform a comparison of the set of points to the model 3D point cloud by comparing the coordinate positions of points within the set of points to coordinate positions of points within the model 3D point cloud; and

based on the performed comparison, identify outlier points within the sets of points representing damage.

5. The system of claim 3, wherein the instructions to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprise instructions to:

superimpose the set of points on the model 3D point cloud;

for each point within the set of points;

identify a nearest neighbor distance for a point within the model 3D point nearest each point; and

analyze the nearest neighbor distance for each point within the set of points to identify outlier points within the set of points representing damage.

6. A computer-implemented method for estimating damage to a roof, the method comprising:

(A) generating, via one or more 3-dimensional (3D) scanners, a 3D point cloud representing a roof;

(B) storing to a memory the 3D point cloud generated via the one or more 3D scanners;

(C) causing one or more processors to retrieve from the memory the 3D point cloud generated via the one or more 3D scanners;

(D) utilizing region growing segmentation to identify from the 3D point cloud a plurality of sets of points, each set of points representing an individual shingle on the roof, including doing the following for each set of points;

(i) identifying a group of points from the 3D point cloud;

(ii) determining a first measurement of a surface shape of the group of points;

(iii) identifying an additional one or more points from the 3D point cloud;

(iv) adding the additional one or more points to the group of points;

(v) determining a second measurement of the surface shape of the group of points; and

(vi) removing the additional one or more points from the group of points when the second measurement of the surface shape exceeds a threshold, wherein the first measurement and second measurement each relate to at least one of;

(a) curvature;

or (b) smoothness;

(E) analyzing, by the one or more processors, each of the plurality of sets of points to identify a set of points representing a damaged shingle; and

(F) generating and displaying a report indicating the damaged shingle is damaged, wherein the displayed report includes one or more of;

(a) a textual representation of damage done to the damaged shingle;

or (b) a graphical representation of damage done to the damaged shingle.

7. The method of claim 6, wherein analyzing each of the plurality of sets of points to identify a set of points representing a damaged shingle comprises:

comparing the set of points to the other sets within the plurality of sets of points to determine that the set of points represents an outlier relative to the other sets.

8. The method of claim 6, wherein analyzing each of the plurality of sets of points to identify a set of points representing a damaged shingle comprises:

generating from the plurality of sets of points a model 3D point cloud representing a model shingle.

9. The method of claim 8, wherein analyzing each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprises:

performing a comparison of the set of points to the model 3D point cloud by comparing the coordinate positions of points within the set of points to coordinate positions of points within the model 3D point cloud; and

based on the performed comparison, identifying outlier points within the sets of points representing damage.

10. The method of claim 8, wherein analyzing each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprises:

superimposing the set of points on the model 3D point cloud;

for each point within the set of points;

identifying a nearest neighbor distance for a point within the model 3D point nearest each point; and

analyzing the nearest neighbor distance for each point within the set of points to identify outlier points within the set of points representing damage.

11. A non-transitory computer readable medium storing instructions that, when executed, cause one or more processors to:

(A) generate a 3D point cloud representing a roof;

(B) retrieve the 3D point cloud;

(C) utilize region growing segmentation to identify from the 3D point cloud a plurality of sets of points, each set of points representing an individual shingle on the roof, including doing the following for each set of points;

(i) identify a group of points from the 3D point cloud;

(ii) determine a first measurement of a surface shape of the group of points;

(iii) identify an additional one or more points from the 3D point cloud;

(iv) add the additional one or more points to the group of points;

(v) determine a second measurement of the surface shape of the group of points; and

(vi) remove the additional one or more points from the group of points when the second measurement of the surface shape exceeds a threshold, wherein the first measurement and second measurement each relate to at least one of;

(i) curvature;

or (ii) smoothness;

(D) analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle; and

(E) generate and display a report indicating the damaged shingle is damaged, wherein the displayed report includes one or more of;

(a) a textual representation of damage done to the damaged shingle;

or (b) a graphical representation of damage done to the damaged shingle.

12. The computer readable medium of claim 11, wherein the instruction to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle comprises instructions to:

compare the set of points to the other sets within the plurality of sets of points to determine that the set of points represents an outlier relative to the other sets.

13. The computer readable medium of claim 11, wherein the instructions to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle comprises instructions to:

generate from the plurality of sets of points a model 3D point cloud representing a model shingle.

14. The computer readable medium of claim 13, wherein the instruction to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprises instructions to:

perform a comparison of the set of points to the model 3D point cloud by comparing the coordinate positions of points within the set of points to coordinate positions of points within the model 3D point cloud; and

identify, based on the performed comparison, outlier points within the sets of points representing damage.

15. The computer readable medium of claim 13, wherein the instruction to analyze each of the plurality of sets of points to identify a set of points representing a damaged shingle further comprises instructions to:

superimpose the set of points on the model 3D point cloud;

identify, for each point within the set of points, a nearest neighbor distance for a point within the model 3D point nearest each point; and

analyze the nearest neighbor distance for each point within the set of points to identify outlier points within the set of points representing damage.