Systems and methods for assessing property damage

• US 8,756,085 B1
• Filed: 10/07/2013
• Issued: 06/17/2014
• 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
• Primary Examiner(s): Poinvil, Frantzy
• Application Number:
  US14/047,844
• Time in Patent Office: 253 Days
• Field of Search: 705 4- 44
• US Class Current: 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 property, estimating damage to the property by analyzing a point cloud of a property. The damage assessment module identifies individual point clusters or segments from the point cloud and detects potentially damaged areas of the property surface by identifying outlier points in the point clusters. The damage assessment module may be used to determine the financial cost of the damage and/or determine whether the property should be replaced or repaired. In addition to eliminating the need for an estimator to visit the property in person, the damage assessment module improves the consistency and accuracy associated with estimating damage to a property.

Claims

1. A computer implemented method for estimating damage to a property, the method comprising:

scanning a property with a 3-dimensional (3D) scanner to generate a first point cloud representing at least part of the property, wherein said property is selected from the group consisting of;

(i) a building, and (ii) a vehicle;

storing the first point cloud to a memory communicatively connected to the 3D scanner;

causing one or more processors, at least one of which is communicatively connected to the memory, to isolate a plurality of segments from the first point cloud, each segment corresponding to one of a plurality of objects detectable at the surface of the property, the plurality of segments including a first segment corresponding to a first object detectable at the surface of the property;

causing one or more processors to generate a model point cloud representing an average of a plurality of point clusters;

causing said one or more processors to compare the first segment to the model point cloud, wherein said comparison comprises causing said one or more processors to;

identify a first point in the first segment,

determine a first coordinate position of the first point,

identify a second point in the model point cloud, wherein the second point is closer to the first coordinate position than any other point in the model point cloud,

determine a second coordinate position of the second point,

determine a distance between the second coordinate position and the first coordinate position, and

determine whether the first point is an outlier point based on the determined distance;

causing said one or more processors to identify, based at least in part on the comparison of the first segment to the model point cloud, a set of outlier points forming a first point cluster and associating the set of outlier points to a damage status; and

causing the one or more processors to generate a display providing information identifying potential damage to the property based at least in part on the damage status.

2. The method of claim 1, wherein:

the property is a building; and

the plurality of objects on the surface of the property is selected from the group consisting of;

(i) a plurality of tiles;

(ii) a plurality of shingles;

(iii) a plurality of bricks;

(iv) a plurality of siding panels; and

(v) a plurality of wood planks.

3. The method of claim 1, wherein:

the property is a vehicle; and

the plurality of objects on the surface of the property is a plurality of body panels.

4. The method of claim 1, wherein causing one or more processors to isolate a plurality of segments from the first point cloud comprises detecting a plurality of point clusters, each point cluster forming a planar region corresponding to one of the plurality of segments.

5. The method of claim 1, wherein causing the one or more processors to isolate a plurality of segments from the first point cloud comprises causing the one or more processors to utilize region growing segmentation to detect a plurality of point clusters forming each of the segments.

6. The method of claim 5, wherein causing the one or more processors to utilize region growing segmentation comprises:

causing the one or more processors to identify a set of candidate points, the set of candidate points including one or more first points from the first point cloud;

causing the one or more processors to determine a first measurement of a surface shape associated with the set of candidate points;

causing the one or more processors to identify one or more second points from the first point cloud;

causing the one or more processors to add the one or more second points to the set of candidate points;

causing the one or more processors to remove the one or more second points from the set of candidate points when a second measurement of a surface shape associated with the set of candidate points exceeds a threshold shape-measurement; and

storing the set of candidate points to a memory as the first segment.

7. The method of claim 6, wherein the first measurement, the second measurement, and the threshold shape-measurement each relate to at least one of:

(i) curvature, or (ii) smoothness.

8. The method of claim 1, wherein the model point cloud includes points from each of a plurality of segments, the plurality of segments including the first segment.

9. The method of claim 1, wherein causing the one or more processors to generate a model point cloud representing an average of a group includes causing the one or more processors to utilize an iterative closest point method to align each of the point clouds in the group of point clouds.

10. The method of claim 1, wherein determining whether the first point is an outlier point based on the determined distance comprises:

causing the one or more processors to determine an average nearest-neighbor distance associated with the model point cloud;

causing the one or more processors to calculate a threshold nearest-neighbor distance based on the determined average nearest-neighbor distance; and

causing the one or more processors to categorize the first point as an outlier point when the determined distance exceeds the threshold nearest-neighbor distance.

11. The method of claim 1, wherein identifying a set of outlier points forming a first point cluster comprises causing one or more processors to utilize Euclidean clustering to identify the set of outlier points forming the first point cluster.

12. The method of claim 1, wherein causing the one or more processors to generate a model point cloud representing an average of a plurality of point clusters comprises:

causing the one or more processors to utilize an iterative closest point method to align each of the point clusters; and

causing the one or more processors to identify a set of points as the model point cloud, the set of points including at least one point from each of the plurality of point clusters.

13. The method of claim 1, wherein the damage status represents at least one of:

(i) a false positive associated with the identified set of outlier points, the false positive indicating that the set of outlier points represent an undamaged area of the object;

or (ii) an extent of damage associated with the identified set of outlier points.

14. The method of claim 1, further including causing the one or more processors to map the set of outlier points to the first point cloud.

15. A system for determining whether an object is damaged beyond expected wear and tear, the system comprising:

one or more 3-dimensional (3D) scanners for scanning a group of two or more objects to generate a group of two or more point clusters, the group of two or more point clusters including a first point cluster representing the first object and a second point cluster representing the second object, wherein the first object and the second object are each components of;

(i) a building, or (ii) a vehicle;

one or more memories for receiving and storing the group of two or more point clusters generated at the one or more 3D scanners;

one or more processors coupled to the one or more memories, wherein the one or more memories include non-transitory computer readable instructions that, when executed at the one or more processors, cause the one or more processors to;

generate a model point cloud including at least one point from each of the two or more point clusters in the group;

identify;

a first nearest-neighbor distance between a first point in the model point cloud and a second point in the model point cloud, the second point being closer to the first point than any other point in the model point cloud, and

a second nearest-neighbor distance between a third point in the model point cloud and a fourth point in the model point cloud, the fourth point being closer to the third point than any other point in the model point cloud;

calculate a threshold nearest-neighbor distance based on an average nearest-neighbor distance, the average nearest-neighbor distance determined, at least in part, based on the first nearest-neighbor distance and the second nearest-neighbor distance;

compare the first point cluster to the model point cloud, wherein said compare comprises determining a third nearest-neighbor distance for a particular point in the first point cluster, the third nearest-neighbor distance being between the particular point in the first point cluster and a nearest point in the model point cloud that is nearest, relative to other points in the model point cloud, to the particular point in the first point cluster;

determine whether the particular point represents a point of potential damage to the first object, exceeding expected wear and tear, by comparing the third nearest-neighbor distance to the threshold nearest-neighbor distance; and

generate a display indicating that the first object is potentially damaged beyond expected wear and tear when it is determined that the particular point represents a point of potential damage to the first object exceeding expected wear and tear.

16. The computer system of claim 15, wherein the instructions that cause the one or more processors to generate a model point cloud including at least one point from each of the two or more point clusters in the group further comprise instructions that cause one or more processors to utilize region growing segmentation to identify each of the two or more point clusters.

17. The computer system of claim 15, wherein the instructions that cause the one or more processors to generate a model point cloud including at least one point from each of the two or more point clusters in the group further comprise instructions that cause the one or more processors to utilize an iterative closest point method to align each of the two or more point clusters in the group.

18. The computer system of claim 15, wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the one or more processors to determine a financial cost associated with the potential damage to the first object when it is determined that the particular point represents a point of potential damage.

19. The computer system of claim 18, wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the one or more processors to initiate a payment based on the determined financial cost.

20. The computer system of claim 15, wherein the first object and the second object are each components of a vehicle.

21. The system of claim 20, wherein the vehicle is:

(i) an automobile, (ii) a watercraft, or (iii) an aircraft.