Systems and methods for rapid three-dimensional modeling with real facade texture

• US 9,836,882 B2
• Filed: 12/09/2016
• Issued: 12/05/2017
• Est. Priority Date: 12/03/2007
• Status: Active Grant

• Current Assignee: Pictometry International Corporation (Eagleview Technologies, Inc. /Wa/)
• Original Assignee: Pictometry International Corporation (Eagleview Technologies, Inc. /Wa/)
• Inventors: Freund, Joseph G., Gal, Ran
• Primary Examiner(s): Hoang, Phi
• Application Number:
  US15/374,358
• Publication Number:
  US 20170193694A1
• Time in Patent Office: 361 Days
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: G06T 15/04 Texture mapping
• CPC Class Codes: G06T 17/05 Geographic models

Abstract

A computer system is described for automatically generating a 3D model, including hardware and non-transitory computer readable medium accessible by the hardware and storing instructions that when executed by the hardware cause it to obtain wire-frame data of a structure within a geographic area; identify a geographic location of the structure; retrieve multiple geo-referenced oblique images representing the geographic location and containing a real façade texture of the structure; locate a geographical position of a real façade texture of the structure; select one or more base oblique image from the multiple oblique images by analyzing, with selection logic, image raster content of the real façade texture depicted in the multiple oblique images, and, apply the real façade texture of the one or more base oblique image to the wire-frame data of the structure to create a three dimensional model providing a real-life representation of physical characteristics of the structure.

Claims

1. A computer system for automatically generating a three-dimensional model, comprising:

hardware;

one or more computer server storing geo-referenced oblique images; and

one or more non-transitory computer readable medium accessible by the hardware and storing instructions that when executed by the hardware cause the hardware to;

obtain wire-frame data of a structure having a roof within a geographical area;

identify a geographic location of the roof of the structure within the geographical area;

retrieve from the one or more computer server multiple geo-referenced oblique images representing the geographic location and containing one or more real faç

ade texture of the roof of the structure;

locate a geographical position of one or more real faç

ade texture of the roof of the structure;

select one or more base oblique image from the multiple geo-referenced oblique images by analyzing, with selection logic, image raster content of the real faç

ade texture depicted in the multiple geo-referenced oblique images, the selection logic using a factorial analysis of the image raster content; and

,

apply the real faç

ade texture of the one or more base oblique image to the wire-frame data of the structure to create a three-dimensional model providing a real-life representation of physical characteristics of the roof of the structure.

2. The computer system of claim 1, wherein locating a geographical position of one or more real faç

ade texture of the roof of the structure comprises locating the geographical position of at least one real faç

ade texture of the roof of the structure utilizing the wire-frame data.

3. The computer system of claim 1, wherein locating a geographical position of one or more real faç

ade texture of the roof of the structure comprises locating the geographical position of at least one real faç

ade texture of the roof of the structure utilizing terrain data.

4. The computer system of claim 1, wherein the factorial analysis is a weighted determination based on at least two factors, the factors including at least one of resolution of the oblique images, colour depth of the oblique images, proportional size of the real faç

ade texture of the oblique images, contrast in the oblique images, time of day the oblique images were captured, time of year the oblique images were captured, amount of foliage in the oblique images, and contrast in lighting conditions in the oblique images.

5. The computer system of claim 1, wherein the one or more non-transitory computer readable medium stores instructions that when executed by the hardware further causes the hardware to:

analyze the real faç

ade texture depicted within the base oblique image to locate a geographical position of at least one occlusion;

locate at least one oblique image having an unoccluded image characteristic of the occlusion in the real faç

ade texture from the multiple oblique images; and

,

apply the unoccluded image characteristic to the real faç

ade texture forming a mosaic image.

6. The computer system of claim 5, wherein the instructions are configured to use pixel pattern recognition of the real faç

ade texture to locate the occlusion.

7. The computer system of claim 5, wherein the oblique images are geo-referenced solved oblique images.

8. The computer system of claim 5, wherein the instructions are configured to remove the occlusion from the real faç

ade texture.

9. The computer system of claim 1, further comprising instructions that when executed by the hardware cause the hardware to extract the real faç

ade texture from the base oblique image, the real faç

ade texture having geo-referenced information.

10. The computer system of claim 1, wherein the real faç

ade texture includes overlapping structures having a geometry, and wherein the instructions, when executed by the hardware cause the hardware to analyze the geometry of the overlapping structures to determine whether the overlapping structures are occlusions or desirable features of the real faç

ade texture.

11. The computer system of claim 10, wherein a first overlapping structure of the overlapping structures is a roof structure.

12. The computer system of claim 10, wherein the one or more non-transitory computer readable medium stores instructions that when executed by the hardware further causes the hardware to:

apply the real faç

ade texture of the overlapping structures that are determined to be desirable features to the three-dimensional model.

13. A method for automatically generating a three-dimensional model, comprising:

obtaining, using one or more computer processor, wire-frame data of at least one structure having a roof;

identifying, using the one or more computer processor, a geographic location of the roof of the structure;

retrieving, from one or more computer server, multiple geo-referenced oblique images representing the geographic location and containing one or more real faç

ade texture of the roof of the structure;

locating, using the one or more computer processor, a geographical position of one or more real faç

ade texture of the roof of the structure;

selecting, using the one or more computer processor, one or more base oblique image from the multiple geo-referenced oblique images by analyzing, with selection logic, image raster content of the real faç

ade texture depicted in the multiple oblique images, the selection logic using a factorial analysis of the image raster content; and

,

applying, using the one or more computer processor, the real faç

ade texture of the one or more base oblique image to the wire-frame data of the structure to create a three-dimensional model providing a real-life representation of physical characteristics of the roof of the structure.

14. The method of claim 13, wherein locating a geographical position of one or more real faç

ade texture of the structure comprises locating the geographical position of at least one real faç

ade texture of the roof of the structure utilizing the wire-frame data.

15. The method of claim 13, wherein locating a geographical position of one or more real faç

ade texture of the roof of the structure comprises locating the geographical position of at least one real faç

ade texture of the structure utilizing terrain data.

16. The method of claim 13, wherein the factorial analysis is a weighted determination based on at least two factors, the factors including at least one of resolution of the oblique images, colour depth of the oblique images, proportional size of the real faç

ade texture of the oblique images, contrast in the oblique images, time of day the oblique images were captured, time of year the oblique images were captured, amount of foliage in the oblique images, and contrast in lighting conditions in the oblique images.

17. The method of claim 13, further comprising:

analyzing, using the one or more computer processor, the real faç

ade texture depicted within the base oblique image to locate a geographical position of at least one occlusion;

locating, using the one or more computer processor, at least one oblique image having an unoccluded image characteristic of the occlusion in the real faç

ade texture from the multiple oblique images; and

,

applying, using the one or more computer processor, the unoccluded image characteristic to the real faç

ade texture forming a mosaic image.

18. The method of claim 17, further comprising removing, using the one or more computer processor, the occlusion from the real faç

ade texture.

19. The method of claim 13, wherein the real faç

ade texture includes overlapping structures having a geometry, and further comprising, analyzing, using the one or more computer processor, the geometry of the overlapping structures to determine whether the overlapping structures are occlusions or desirable features of the real faç

ade texture.

20. The method of claim 19, further comprising:

applying, using the one or more computer processor, the real faç

ade texture of the overlapping structures that are determined to be desirable features to the three-dimensional model.