IMAGE SEGMENTATION SYSTEM FOR VERIFICATION OF PROPERTY ROOF DAMAGE

• US 20170352100A1
• Filed: 06/03/2016
• Published: 12/07/2017
• Est. Priority Date: 06/03/2016
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: Shreve, Matthew Adam, Bernal, Edgar A., Howe, Richard L.
• Granted Patent: US 11,055,786 B2
• Time in Patent Office: Days
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G06T 2207/10032 Satellite or aerial image; Remote sensing
• CPC Class Codes: G06T 2207/20061 Hough transform
• CPC Class Codes: G06T 2207/30184 Infrastructure
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/168 involving transform domain methods
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/753 Transform-based matching, e.g. Hough transform
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

A system segments a set of images of a property to identify a type of damage to the property. The system receives, from an image capturing device, a digital image of a roof or other feature of the property. The system processes the image to identify a set of segments, in which each segment corresponds to a piece of the feature, such as a tab or tooth of a shingle on the roof. The system saves a result of the processing to a data file as a segmented image of the property, and it uses the segmented image to identify a type of damage to the property.

Claims

1. A system for processing an image to segment the image for property damage assessment, the system comprising:

a processing device; and

a computer-readable medium containing programming instructions that are configured to cause the processing device to;

receive, from an image capturing device, a digital image of a roof of a property,

process the image to identify a plurality of segments, each of which corresponds to a tab or tooth of a shingle on the roof,

save a result of the processing to a data file as a segmented image of the roof, and

use the segmented image to identify a type of damage to the roof.

2. The system of claim 1, wherein the instructions to process the digital image to identify a plurality of segments comprise instructions to:

produce an edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

use the boundaries to identify as the segments a plurality of the tabs or teeth for which boundaries are identified.

3. The system of claim 1, wherein the instructions to process the digital image to identify a plurality of segments comprise instructions to:

produce an initial edge map by extracting a plurality of edges from the image, so that at least some of the edges correspond to shingle boundaries;

use a line detector to detect a plurality of lines in the image;

determine a top N orientation of the lines, based on those orientations that occur most frequently;

produce an enhanced edge map by adding pixels that are covered by the lines that are oriented in one of the top N orientations to the initial edge map; and

fill a plurality of cells of the enhanced edge map.

4. The system of claim 1, wherein the instructions to process the digital image to identify a plurality of segments comprise instructions to:

produce an edge map by;

identifying a plurality of edges in the digital image,

identifying a sensitivity threshold T,

applying Canny edge detection to each of the edges with the sensitivity threshold T to identify a subset of edges having a magnitude that is greater than the sensitivity threshold T, and

saving all edges having a magnitude that is greater than the sensitivity threshold T to the edge map.

5. The system of claim 1, wherein the instructions to process the digital image to identify a plurality of segments comprise instructions to:

produce an edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

produce a refined edge map by;

applying a Hough transform to detect a plurality of lines, and

determining a number of highest peaks of the detected lines.

6. The system of claim 5, wherein the instructions to process the digital image to identify a plurality of segments further comprise instructions to enhance the edge map by accepting all lines oriented within radius r around each peak, and labeling each pixel of each line as an edge pixel in the refined edge map.

7. The system of claim 6 wherein the instructions to process the digital image to identify a plurality of segments further comprise instructions to:

assign a unique label to each group of connected non-edge pixels; and

consider each labeled group to be a tab or tooth.

8. A method of processing an image to segment the image for property damage assessment, the method comprising:

by a processing device, executing programming instructions that cause the processing device to;

receive, from an image capturing device, a digital image of a roof of a property,

process the digital image to identify a plurality of segments, each of which corresponds to a tab or tooth of a shingle on the roof,

save a result of the processing to a data file as a segmented image of the roof, and

use the segmented image to identify a type of damage to the roof.

9. The method of claim 8, wherein the processing of the digital image to identify a plurality of segments comprises:

producing an edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

using the boundaries to identify as the segments a plurality of the tabs or teeth for which boundaries are identified.

10. The method of claim 8, wherein the processing of the digital image to identify a plurality of segments comprises:

producing an initial edge map by extracting a plurality of edges from the image, so that at least some of the edges correspond to shingle boundaries;

using a line detector to detect a plurality of lines in the image;

determining a top N orientation of the lines, based on those orientations that occur most frequently;

producing an enhanced edge map by adding pixels that are covered by the lines that are oriented in one of the top N orientations to the initial edge map; and

filling a plurality of cells of the enhanced edge map.

11. The method of claim 8, wherein the processing of the digital image to identify a plurality of segments comprises:

producing an edge map by;

identifying a plurality of edges in the digital image,

identifying a sensitivity threshold T,

applying Canny edge detection to each of the edges with the sensitivity threshold T to identify a subset of edges having a magnitude that is greater than the sensitivity threshold T, and

saving all edges having a magnitude that is greater than the sensitivity threshold T to the edge map.

12. The method of claim 8, wherein the processing of the digital image to identify a plurality of segments comprises:

producing an edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

producing a refined edge map by;

applying a Hough transform to detect a plurality of lines, and

determining a number of highest peaks of the detected lines.

13. The method of claim 12, wherein the processing of the digital image to identify a plurality of segments further comprises:

enhancing the edge map by accepting all lines oriented within radius r around each peak; and

labeling each pixel of each line as an edge pixel in the refined edge map.

14. The method of claim 13 wherein the processing of the digital image to identify a plurality of segments comprises:

assigning a unique label to each group of connected non-edge pixels; and

considering each labeled group to be a tab or tooth.

15. A method of processing an image to segment the image for property damage assessment, the method comprising:

by a processing device, executing programming instructions that cause the processing device to;

receive, from an image capturing device, a digital image of a roof of a property,

process the digital image to produce an edge map of the image and use the edge map to identify a plurality of segments, each of which corresponds to a tab or tooth of a shingle on the roof,

save a result of the processing to a data file as a segmented image of the roof, and

use the segmented image to identify a type of damage to the roof.

16. The method of claim 15, wherein the processing of the digital image to produce the edge map of the image and use the edge map to identify a plurality of segments comprises:

producing the edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

using the boundaries to identify as the segments a plurality of the tabs or teeth for which boundaries are identified.

17. The method of claim 15, wherein the processing of the digital image to produce the edge map of the image and use the edge map to identify a plurality of segments comprises:

producing an initial edge map by extracting a plurality of edges from the image, so that at least some of the edges correspond to shingle boundaries;

using a line detector to detect a plurality of lines in the image;

determining a top N orientation of the lines, based on those orientations that occur most frequently;

producing an enhanced edge map by adding pixels that are covered by the lines that are oriented in one of the top N orientations to the initial edge map; and

filling a plurality of cells of the enhanced edge map.

18. The method of claim 15, wherein the processing of the digital image to produce the edge map of the image and use the edge map to identify a plurality of segments comprises:

producing the edge map by;

identifying a plurality of edges in the digital image,

identifying a sensitivity threshold T,

applying Canny edge detection to each of the edges with the sensitivity threshold T to identify a subset of edges having a magnitude that is greater than the sensitivity threshold T, and

saving all edges having a magnitude that is greater than the sensitivity threshold T to the edge map.

19. The method of claim 15, wherein the processing of the digital image to produce the edge map of the image and use the edge map to identify a plurality of segments comprises:

producing an edge map by extracting a plurality of edges from the digital image so that at least some of the edges correspond to boundaries of a tab or tooth of a shingle; and

producing a refined edge map by;

applying a Hough transform to detect a plurality of lines, and

determining a number of highest peaks of the detected lines.