Texture identification

1. A method to determine a texture of a surface of an object, the method comprising:

receiving an image comprising a two-dimensional (2D) representation of the surface of the object;

classifying the surface into at least one texture type from a plurality of texture types, wherein the texture type is based on a plurality of parameters for the image;

using a wavelet transform to generate a scalogram of the surface, wherein the type of wavelet transform is selected based on the texture type of the surface;

using a Fourier expansion of the scalogram to determine affine invariant features; and

determining the texture of the surface from the affine invariant features.

2. The method of claim 1 , further comprising estimating a three-dimensional (3D) projection of the image.

3. The method of claim 2 , wherein estimating the 3D projection of the image comprises projecting the 2D image onto a Cartesian coordinate system.

4. The method of claim 1 , wherein determining the plurality of parameters for the image comprises:

generating a directional histogram based on the image; and

determining a directionality parameter for the image by calculating a number of peaks in the histogram.

5. The method of claim 1 , wherein the plurality of parameters for the image are determined by:

generating a co-occurrence matrix based on the image; and

determining a homogeneity parameter based on the co-occurrence matrix.

6. The method of claim 1 , wherein the plurality of parameters for the image are determined by:

calculating a root mean square value of the image; and

determining a roughness parameter based on the root mean square value of the image.

7. The method of claim 1 , wherein the wavelet transform is a regular log Gabor wavelet.

8. The method of claim 1 , wherein the wavelet transform is a Daubecheis wavelet.

9. The method of claim 1 , wherein the wavelet transform is a Mexican hat wavelet.

10. The method of claim 1 , wherein the wavelet transform is a Gabor wavelet.

11. The method of claim 1 , wherein the wavelet transform is a directional log Gabor wavelet.

12. A system to determine a texture of a surface of an object, the system comprising:

a processor configured to:

access an image comprising a two-dimensional (2D) representation of the surface of the object;

classify the surface into at least one texture type from a plurality of texture types; wherein the texture type is based on a plurality of parameters for the image;

use a wavelet transform to generate a scalogram of the surface, wherein the type of wavelet transform is selected based on the texture type of the surface;

use a Fourier transform of a scalogram to determine affine invariant features; and

determine the texture of the surface from the affine invariant features.

13. The system of claim 12 , wherein the plurality of parameters is selected from the group consisting of homogeneity, directionality, regularity, and roughness.

14. The system of claim 12 , wherein the plurality of texture types is selected from the group consisting of homogenous texture, directional texture, regular texture, and rough texture.

15. The system of claim 12 , wherein the processor is configured to classify the surface into a single texture type.

16. The system of claim 12 , wherein the processor is further configured to:

estimate a three dimensional (3D) projection of the image;

transform the 3D projection into a frequency domain;

project the 3D projection in the frequency domain on to a spherical co-ordinate system; and

generate a spectral signature from the 3D projection on the spherical co-ordinate system.

17. The system of claim 12 , wherein the wavelet transform is a regular log Gabor wavelet.

18. The system of claim 12 , wherein the wavelet transform is a Daubecheis wavelet.

19. The system of claim 12 , wherein the wavelet transform is a Mexican hat wavelet.

20. The system of claim 12 , wherein the wavelet transform is a Gabor wavelet.

21. The system of claim 12 , wherein the wavelet transform is a directional log Gabor wavelet.

22. A non-transitory computer program product, for use in a computing system including a processor and a memory, for implementing a method for performing texture identification of the surface of an object, the computer program product comprising one or more physical non-transitory computer readable medium having stored thereon computer-executable instructions that, when executed by the processor, causes the computing system to:

receive an image comprising a two-dimensional (2D) representation of the surface of the object;

classify the surface into at least one texture type from a plurality of texture types, wherein the texture type is based on a plurality of parameters for the image;

use a wavelet transform to generate a scalogram of the surface, wherein the type of wavelet transform is selected based on the texture type of the surface;

use a Fourier transform of the scalogram to determine affine invariant features; and

determine the texture of the surface from the affine invariant features.

23. The computer program product of claim 22 , wherein the wavelet transform is a regular log Gabor wavelet.

24. The computer program product of claim 22 , wherein the wavelet transform is a Daubecheis wavelet.

25. The computer program product of claim 22 , wherein the wavelet transform is a Mexican hat wavelet.

26. The computer program product of claim 22 , wherein the wavelet transform is a Gabor wavelet.

27. The computer program product of claim 22 , wherein the wavelet transform is a directional log Gabor wavelet.