Apparatus and method for characterizing texture

• US 9,189,864 B2
• Filed: 02/05/2014
• Issued: 11/17/2015
• Est. Priority Date: 10/17/2013
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Kellomki, Markku
• Primary Examiner(s): Allison, Andrae S
• Application Number:
  US14/173,284
• Publication Number:
  US 20150110359A1
• Time in Patent Office: 650 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 2207/30124 Fabrics; Textile; Paper
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529
• CPC Class Codes: G06V 10/42 Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation

Abstract

A method includes, using at least one processing device, obtaining an image of a surface having a texture and identifying a dominant size of the texture using a discrete auto-covariance function of the image. A first positive local maximum of the discrete auto-covariance function could be identified. The discrete auto-covariance function could include points associated with positive numbers of whole pixels, and the first positive local maximum of the discrete auto-covariance function could be identified at one of the points. Sub-pixel estimation could also be performed using the point associated with the first positive local maximum and one or more neighboring points. Performing the sub-pixel estimation could include fitting a polynomial curve to the point associated with the first positive local maximum and the one or more neighboring points and identifying a number of whole and fractional pixels associated with a maximum of the polynomial curve.

Claims

1. A method comprising:

using at least one processing device;

obtaining a multi-dimensional image of a surface having a texture;

identifying multiple discrete auto-covariance functions of the multi-dimensional image, each discrete auto-covariance function identified using a different set of pixels in the multi-dimensional image;

averaging the multiple discrete auto-covariance functions to generate a final discrete auto-covariance function of the multi-dimensional image; and

identifying a dominant size of the texture using the final discrete auto-covariance function.

2. The method of claim 1, wherein identifying the dominant size of the texture comprises:

identifying a first positive local maximum of the final discrete auto-covariance function.

3. The method of claim 2, wherein:

the final discrete auto-covariance function comprises points associated with positive numbers of whole pixels;

the first positive local maximum of the final discrete auto-covariance function is identified at one of the points; and

identifying the dominant size of the texture further comprises performing sub-pixel estimation using the point associated with the first positive local maximum and one or more neighboring points.

4. The method of claim 3, wherein performing the sub-pixel estimation comprises:

fitting a polynomial curve to the point associated with the first positive local maximum and the one or more neighboring points; and

identifying a number of whole and fractional pixels associated with a maximum of the polynomial curve.

5. The method of claim 4, further comprising:

identifying a scale of the multi-dimensional image, the scale identifying a distance associated with each pixel of the multi-dimensional image; and

multiplying the number of whole and fractional pixels by the scale to identify a length of the dominant size of the texture.

6. The method of claim 1, further comprising:

pre-processing the multi-dimensional image prior to identifying the dominant size of the texture.

7. The method of claim 6, wherein pre-processing the multi-dimensional image comprises adjusting the multi-dimensional image to compensate for uneven illumination of the surface.

8. The method of claim 1, further comprising:

performing one or more optical, geometrical, or statistical corrections of the multi-dimensional image.

9. The method of claim 1, wherein the multi-dimensional image comprises a multi-dimensional image of a surface having a randomized texture.

10. The method of claim 1, wherein the multiple discrete auto-covariance functions of the multi-dimensional image are identified using pixels in different rows, columns, or radial directions of the multi-dimensional image.

11. An apparatus comprising:

at least one memory configured to store a multi-dimensional image of a surface having a texture; and

at least one processing device configured to;

identify multiple discrete auto-covariance functions of the multi-dimensional image, each discrete auto-covariance function identified using a different set of pixels in the multi-dimensional image;

average the multiple discrete auto-covariance functions to generate a final discrete auto-covariance function of the multi-dimensional image; and

identify a dominant size of the texture using the final discrete auto-covariance function.

12. The apparatus of claim 11, wherein the at least one processing device is configured to identify the dominant size of the texture by identifying a first positive local maximum of the final discrete auto-covariance function.

13. The apparatus of claim 12, wherein:

the final discrete auto-covariance function comprises points associated with positive numbers of whole pixels;

the at least one processing device is configured to identify the first positive local maximum of the final discrete auto-covariance function at one of the points; and

the at least one processing device is configured to identify the dominant size of the texture by performing sub-pixel estimation using the point associated with the first positive local maximum and one or more neighboring points.

14. The apparatus of claim 13, wherein the at least one processing device is configured to perform the sub-pixel estimation by:

fitting a polynomial curve to the point associated with the first positive local maximum and the one or more neighboring points; and

identifying a number of whole and fractional pixels associated with a maximum of the polynomial curve.

15. The apparatus of claim 14, wherein the at least one processing device is further configured to:

identify a scale of the multi-dimensional image, the scale identifying a distance associated with each pixel of the multi-dimensional image; and

multiply the number of whole and fractional pixels by the scale to identify a length of the dominant size of the texture.

16. The apparatus of claim 11, wherein the at least one processing device is further configured to pre-process the multi-dimensional image prior to identifying the dominant size of the texture.

17. The apparatus of claim 16, wherein the at least one processing device is configured to pre-process the multi-dimensional image to compensate for uneven illumination of the surface.

18. The apparatus of claim 11, wherein the at least one processing device is further configured to perform one or more optical, geometrical, or statistical corrections of the multi-dimensional image.

19. The apparatus of claim 11, wherein the multi-dimensional image comprises a multi-dimensional image of a surface having a randomized texture.

20. The apparatus of claim 11, wherein the at least one processing device is configured to identify the multiple discrete auto-covariance functions of the multi-dimensional image using pixels in different rows, columns, or radial directions of the multi-dimensional image.

21. A non-transitory computer readable medium containing a computer program, the computer program comprising computer readable program code that when executed causes at least one processing device to:

obtain a multi-dimensional image of a surface having a texture;

identify multiple discrete auto-covariance functions of the multi-dimensional image, each discrete auto-covariance function identified using a different set of pixels in the multi-dimensional image;

average the multiple discrete auto-covariance functions to generate a final discrete auto-covariance function of the multi-dimensional image; and

identify a dominant size of the texture using the final discrete auto-covariance function.

22. The non-transitory computer readable medium of claim 21, wherein the computer readable program code that when executed causes the at least one processing device to identify the dominant size of the texture comprises computer readable program code that when executed causes the at least one processing device to:

identify a first positive local maximum of the final discrete auto-covariance function, the final discrete auto-covariance function comprising points associated with positive numbers of whole pixels, the first positive local maximum of the final discrete auto-covariance function identified at one of the points; and

perform sub-pixel estimation using the point associated with the first positive local maximum and one or more neighboring points.

23. The non-transitory computer readable medium of claim 22, wherein the computer readable program code that when executed causes the at least one processing device to perform the sub-pixel estimation comprises computer readable program code that when executed causes the at least one processing device to:

fit a polynomial curve to the point associated with the first positive local maximum and the one or more neighboring points; and

identify a number of whole and fractional pixels associated with a maximum of the polynomial curve.

24. The non-transitory computer readable medium of claim 23, wherein the computer program further comprises computer readable program code that when executed causes the at least one processing device to:

identify a scale of the multi-dimensional image, the scale identifying a distance associated with each pixel of the multi-dimensional image; and

multiply the number of whole and fractional pixels by the scale to identify a length of the dominant size of the texture.

25. The non-transitory computer readable medium of claim 21, wherein the computer program further comprises computer readable program code that when executed causes the at least one processing device to:

pre-process the multi-dimensional image prior to identifying the dominant size of the texture.

26. The non-transitory computer readable medium of claim 21, wherein the computer program further comprises computer readable program code that when executed causes the at least one processing device to:

pre-process the multi-dimensional image to compensate for uneven illumination of the surface.

27. The non-transitory computer readable medium of claim 21, wherein the multi-dimensional image comprises a multi-dimensional image of a surface having a randomized texture.

28. The non-transitory computer readable medium of claim 21, wherein the computer readable program code that when executed causes the at least one processing device to identify the multiple discrete auto-covariance functions comprises computer readable program code that when executed causes the at least one processing device to:

identify the multiple discrete auto-covariance functions of the multi-dimensional image using pixels in different rows, columns, or radial directions of the multi-dimensional image.