Building footprint extraction apparatus, method and computer program product

• US 9,639,757 B2
• Filed: 09/23/2011
• Issued: 05/02/2017
• Est. Priority Date: 09/23/2011
• Status: Active Grant

• Current Assignee: Corelogic Solutions LLC (CoreLogic Incorporated)
• Original Assignee: Corelogic Solutions LLC (CoreLogic Incorporated)
• Inventors: Du, Wei, Jeffery, Thomas C., Botts, Howard
• Primary Examiner(s): Kholdebarin, Iman K
• Application Number:
  US13/243,405
• Publication Number:
  US 20130077819A1
• Time in Patent Office: 2,048 Days
• Field of Search: 382100, 382103, 382190
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 20/176 Urban or other man-made structures
• CPC Class Codes: Y02A 10/40 Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping

Abstract

A system, method and computer program product cooperate to extract a building footprint from other data associated with a property. Imagery data of real property is input to a computing device, the imagery data containing a plurality of parcels. A processing circuit detects contrasts of candidate man-made structures on a parcel of the plurality of parcels. The candidate man-made structures are then associated with the parcel. A building footprint is then extracted by distinguishing a man-made structure on said parcel from natural terrain, recognizing that man-made structures when viewed from above generally show a strong contrast from background terrain. Remaining candidate man-made structures are removed by observing that they having features inconsistent with predetermined extraction logic.

Claims

1. A method for extracting a building footprint from imagery data, comprising:

accessing imagery data of land containing a plurality of parcels;

detecting with a processing circuit image contrasts between natural terrain and potential man-made structures in said imagery data;

determining a parcel from the plurality of parcels that a portion of the potential man-made structures are located on by overlaying parcel boundaries over said imagery data so as to spatially distinguish said plurality of parcels; and

extracting a building footprint of a detected man-made structure on said determined parcel from the plurality of parcels that the portion of the potential man-made structures are located on, said extracting including filtering-out remaining potential man-made structures having features inconsistent with predetermined extraction rules defined based at least in part on property characteristics for the determined parcel from the plurality of parcels that the portion of the potential man-made structures are located on.

2. The method of claim 1, wherein,

said imagery data being aerial imagery data.

3. The method of claim 1, wherein,

said extracting includes comparing a building shape on a tax map with the building footprint, and correcting abnormalities in said building footprint with features taken from said building shape on said tax map.

4. The method of claim 1, further comprising:

assigning a property point to a building footprint boundary point; and

determining with a geocode mechanism geocodes that include latitude and longitude for said parcel based on the property point as an output of the geocode mechanism that corresponds to property address information for said parcel.

5. The method of claim 1, further comprising:

assigning a property point to a building footprint centroid; and

determining with a geocode mechanism geocodes that include latitude and longitude for said parcel based on the property point as an output of the geocode mechanism that corresponds to property address information for said parcel.

6. The method of claim 4, wherein

said assigning includes assigning a plurality of property points to said building footprint and determining geocodes for each of said plurality of property points that include multiple points on a single structure and multiple structures within the parcel.

7. The method of claim 5, wherein

said assigning includes assigning a plurality of property points to said building footprint and determining geocodes for each of said plurality of property points that include multiple points on a single structure and multiple structures within the parcel.

8. The method of claim 1, further comprising:

populating a database with property points for each parcel of the plurality of parcels.

9. The method of claim 8, wherein

said populating includes storing said building footprint and property points in association with a parcel description for each of said plurality of parcels.

10. The method of claim 9, wherein

said populating includes populating a national database.

11. The method of claim 1, wherein

said detecting includes adjusting an image threshold of said imagery data to emphasize contrasts between the potential man-made structures and natural terrain.

12. The method of claim 1, wherein

said extracting includes forming polygons for said potential man-made structures, and

said predetermined extraction rules including excluding any polygon having a square footage larger than a recorded square footage of a building on said parcel.

13. The method of claim 12, wherein

said predetermined extraction rules retain a polygon having a square footage closest to the recorded square footage of the building on the parcel.

14. The method of claim 12, wherein

said predetermined extraction rules remove any polygon having a dimension narrower than a predetermined amount.

15. The method of claim 12, wherein

said predetermined extraction rules remove any polygon having a length longer than a predetermined distance.

16. The method of claim 12, wherein

said predetermined extraction rules remove any polygon that extends by at least a predetermined amount beyond a property boundary of said parcel.

17. The method of claim 12, wherein

said predetermined extraction rules remove any polygon that has a complex geometry having more than a predetermined number of sides.

18. The method of claim 1, further comprising:

recording said building footprint in a database; and

repeating said inputting, detecting, associating and extracting steps for other imagery data so as to populate the database with building footprints for parcels spread over a region.

21. The method of claim 1, wherein

said property characteristics comprise a recorded square footage of a building on said parcel.

22. The method of claim 1, further comprising:

assigning a plurality of property points to said building footprint; and

assigning flood risk scores to each of the plurality of property points.

23. The method of claim 22, wherein the flood risk scores are calculated based at least on an elevation associated with each of the plurality of property points and proximity of the plurality of property points to one or more potential flood sources.

19. An apparatus for extracting a building footprint from imagery data, comprising:

an interface that accesses imagery data of land; and

a processing circuit configured to

detect image contrasts between natural terrain and potential man-made structures in said imagery data,

determine a parcel that a portion of the potential man-made structures are located on by overlaying parcel boundaries over said imagery data, and

extract a building footprint of a detected man-made structure on said determined parcel that the portion of the potential man-made structures are located on, including filtering-out remaining potential man-made structures having features inconsistent with predetermined extraction rules defined based at least in part on property characteristics for the determined parcel that the portion of the potential man-made structures are located on.

20. A non-transitory computer program product having instructions that when executed by a processing circuit perform a method of extracting a building footprint from imagery data, comprising:

accessing imagery data of;

detecting with the processing circuit image contrasts between natural terrain and potential man-made structures in said imagery data;

determining a parcel that a portion of the potential man-made structures are located on by overlaying parcel boundaries over said imagery data; and

extracting a building footprint of a detected man-made structure on said determined parcel that the portion of the potential man-made structures are located on, said extracting including filtering-out remaining potential man-made structures having features inconsistent with predetermined extraction rules defined based at least in part on parcel characteristics for the determined parcel that the portion of the potential man-made structures are located on.