System and method for creating a high resolution material image

• US 20070036467A1
• Filed: 05/04/2005
• Published: 02/15/2007
• Est. Priority Date: 07/26/2004
• Status: Abandoned Application

• Current Assignee: Brian Mcdonald, Christopher Coleman
• Original Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Inventors: Coleman, Christopher, McDonald, Brian
• Application Number:
  US11/122,406
• Publication Number:
  US 20070036467A1
• Time in Patent Office: Days
• US Class Current: 382/294
• CPC Class Codes: G06T 15/04 Texture mapping
• CPC Class Codes: G06T 3/4061 by injecting details from different spectral ranges

Abstract

For example, the system may include or execute software for material classification of an image. The software typically comprises computer-readable instructions and is operable to identify a first image of a first type with a first resolution, the first type comprising a visible image. The software is further operable to identify a second image of a second type with a second resolution, with the second image spatially correlated with the first image and the second type comprising a material image. The software then generates a third image of the second type with the first resolution using the first and second images.

Claims

1. Software for material classification of an image, the software comprising computer-readable instructions and operable to:

identify a first image of a first type with a first resolution, the first type comprising a visible image;

identify a second image of a second type with a second resolution, the second image spatially correlated with the first image, the second type comprising a material classification image; and

generate a third image of the second type with the first resolution using the first and second images.

2. The software of claim 1, the first image comprising a higher resolution photographic image and the second image comprising a lower resolution land-cover classification image.

3. The software of claim 2, the higher resolution photographic image having less than or equal to one meter resolution and the lower resolution land-cover classification image having thirty meter resolution.

4. The software of claim 1, wherein the software operable to generate the third image comprises software operable to:

select a first pixel from the second image;

identify a second pixel from the first image that is spatially correlated with the first pixel;

identify a sub-material using, at least in part, a material identification from the first pixel and a reflectance of the second pixel; and

generate a third pixel in the third image based on the identified sub-material.

5. The software of claim 4, wherein the software operable to identify the sub-material comprises software operable to:

determine a gross-classification value using the material identification and the reflectance; and

select a fine-classification based on the gross-classification value, the fine-classification comprising a sub-material identifier in a materials table.

6. The software of claim 4, wherein the software operable to identify the sub-material comprises software operable to:

determine a first reflectance value of the second pixel;

identify a sub-material that has a second reflectance closest to the determined first reflectance value;

determine a difference value between the first and second reflectance values; and

generate a fourth pixel in a difference image using the difference value.

7. The software of claim 1, further operable to:

tile the first image into a virtual texture; and

tile the second image such that coverage of the second image'"'"'s tiles substantially matches that of the first image'"'"'s tiles.

8. The software of claim 1, the second image comprising pixels associated with one of eight materials.

9. A method for material classification of an image, comprising:

identifying a first image of a first type with a first resolution, the first type comprising a visible image;

identifying a second image of a second type with a second resolution, the second image spatially correlated with the first image, the second type comprising a material classification image; and

generating a third image of the second type with the first resolution using the first and second images.

10. The method of claim 9, the first image comprising a higher resolution photographic image and the second image comprising a lower resolution land-cover classification image.

11. The method of claim 10, the higher resolution photographic image having less than or equal to one meter resolution and the lower resolution land-cover classification image having thirty meter resolution.

12. The method of claim 9, wherein generating the third image comprises:

selecting a first pixel from the second image;

identifying a second pixel from the first image that is spatially correlated with the first pixel;

identifying a sub-material using, at least in part, a material identification from the first pixel and a reflectance of the second pixel; and

generating a third pixel in the third image based on the identified sub-material.

13. The method of claim 12, wherein identifying the sub-material comprises:

determining a gross-classification value using the material identification and the reflectance; and

selecting a fine-classification based on the gross-classification value, the fine-classification comprising a sub-material identifier in a materials table.

14. The method of claim 12, wherein identifying the sub-material comprises:

determining a first reflectance value of the second pixel;

identifying a sub-material that has a second reflectance closest to the determined first reflectance value;

determining a difference value between the first and second reflectance values; and

generating a fourth pixel in a difference image using the difference value.

15. The method of claim 9, further comprising:

tiling the first image into a virtual texture; and

tiling the second image such that coverage of the second image'"'"'s tiles substantially matches that of the first image'"'"'s tiles.

16. A system for material classification of an image, the system comprising:

memory storing a plurality of first images of a first type with a first resolution and a plurality of second images of a second type with a second resolution, the first type comprising a visible image and the second type and comprising a material classification image; and

one or more processors operable to;

identify one of the first images of the first type;

identify one of the second images of the second type with a second resolution, the identified second image spatially correlated with the identified first image; and

generate a third image of the second type with the first resolution using the first and second images.

17. The system of claim 16, the first image comprising a higher resolution photographic image and the second image comprising a lower resolution land-cover classification image.

18. The system of claim 17, the higher resolution photographic image having less than or equal to one meter resolution and the lower resolution land-cover classification image having thirty meter resolution.

19. The system of claim 16, wherein the one or more processors operable to generate the third image comprise one or more processors operable to:

select a first pixel from the second image;

identify a second pixel from the first image that is spatially correlated with the first pixel;

identify a sub-material using, at least in part, a material identification from the first pixel and a reflectance of the second pixel; and

generate a third pixel in the third image based on the identified sub-material.

20. The system of claim 19, wherein the one or more processors operable to identify the sub-material comprise one or more processors operable to:

determine a gross-classification value using the material identification and the reflectance; and

select a fine-classification based on the gross-classification value, the fine-classification comprising a sub-material identifier in a materials table.

21. The system of claim 19, wherein the one or more processors operable to identify the sub-material comprise one or more processors operable to:

determine a first reflectance value of the second pixel;

identify a sub-material that has a second reflectance closest to the determined first reflectance value;

determine a difference value between the first and second reflectance values; and

generate a fourth pixel in a difference image using the difference value.

22. The system of claim 16, the one or more processors further operable to:

tile the first image into a virtual texture; and

tile the second image such that coverage of the second image'"'"'s tiles substantially matches that of the first image'"'"'s tiles.