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

• US 20090141020A1
• Filed: 12/03/2007
• Published: 06/04/2009
• 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
• Granted Patent: US 8,531,472 B2
• Time in Patent Office: Days
• US Class Current: 345/419
• 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

Methods are described for automatically generating a three-dimensional model of a structure including real façade textures obtained from geo-referenced oblique images. The geographical position of at least one real façade texture is located. Geo-referenced images containing the real façade texture are located. A base image having the real façade texture is selected from the geo-referenced images by analyzing image raster content. The real façade texture of the base image is applied to the three-dimensional model. Systems for generating the three-dimensional model are also described herein.

Claims

1. A method for automatically generating a three-dimensional model of a structure including real faç

ade textures obtained from geo-referenced oblique images, comprising the steps of;

analyzing wire-frame data of the three-dimensional model of the structure to locate geographical positions of at least one real faç

ade texture of the structure;

locating oblique images containing the real faç

ade texture;

selecting a base oblique image from the oblique images by analyzing image raster content of the real faç

ade texture of each of the oblique images; and

,

applying the real faç

ade texture of the base oblique image to the three-dimensional model.

2. The method of claim 1, further comprising the steps of:

analyzing the real faç

ade texture to locate geographical positions of at least one occlusion;

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

ade texture; and

,

applying the unoccluded image characteristic to the real faç

ade texture forming a mosaic image.

3. The method of claim 2, wherein the occlusion is located using pixel pattern recognition of the real faç

ade texture.

4. The method of claim 2, wherein the oblique images are geo-referenced solved oblique images.

5. The method of claim 2, further comprising the step of removing the occlusion from the real faç

ade texture.

6. The method of claim 1, further comprising the step of extracting the real faç

ade texture from the base oblique image, the real faç

ade texture having geo-referenced information.

7. The method of claim 1, further comprising the steps of:

analyzing wire-frame data to locate the geographical position of a roof of the structure;

locating images containing the roof;

selecting a base image from the images containing the roof by analyzing image raster content; and

,

applying the roof to the three-dimensional model.

8. The method of claim 7, wherein the images containing the roof are nadir images.

9. The method of claim 7, wherein the images containing the roof are oblique images.

10. The method of claim 7, wherein the images containing the roof are nadir images and oblique images.

11. The method of claim 7, wherein the base image contains a greater proportion of pixels associated with the roof as compared with other oblique images.

12. The method of claim 1, wherein the oblique images are geo-referenced solved oblique images.

13. The method of claim 1, further comprising the step of correcting misalignment of the real faç

ade texture.

14. The method of claim 13, wherein the misalignment is corrected using an edge-detection algorithm.

15. The method of claim 1, further comprising the step of aligning the real faç

ade texture to the wire-frame data of the three-dimensional model.

16. The method of claim 1, further comprising the step of aligning the wire-frame data to the faç

ade.

17. A method for automatically generating a three-dimensional model of structures including real faç

ade textures obtained from geo-referenced aerial oblique images, comprising the steps of;

analyzing wire-frame data to locate geographical positions of real faç

ade textures of the structures within a geographical area;

choosing an oblique image showing the real faç

ade texture;

identifying an incorrect area within the part of the real faç

ade texture;

automatically correcting the incorrect area within the part of the real faç

ade texture.