Methods for processing color image data employing a chroma, hue, and intensity color representation

• US 7,358,976 B2
• Filed: 03/25/2004
• Issued: 04/15/2008
• Est. Priority Date: 03/25/2003
• Status: Expired due to Fees

• Current Assignee: VideoIQ, Inc. (Motorola Solutions, Inc.)
• Original Assignee: VideoIQ, Inc. (Motorola Solutions, Inc.)
• Inventors: Saptharishi, Mahesh, Hampshire, John B. II
• Primary Examiner(s): Tung; Kee M.
• Assistant Examiner(s): Amini; Javid A
• Application Number:
  US10/811,376
• Publication Number:
  US 20040234126A1
• Time in Patent Office: 1,482 Days
• Field of Search: 345/589, 382/162
• US Class Current: 345/589
• CPC Class Codes: G01J 3/46 Measurement of colour; Colour measuring devices, e.g. colorimeters measuring colour temperature G01J5/60
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: H04N 1/646 Transmitting or storing colour television type signals, e.g. PAL, Lab; Their conversion into additive or subtractive colour signals or vice versa therefor H04N1/642, H04N1/644 take precedence

Abstract

A method senses electromagnetic energy associated with a source over an area in N frequency bands and generates color image data representing at least a portion of the area. The data are arranged as pixels, and the data for a given pixel comprise chroma, hue, and intensity values. The N frequency bands constitute a mathematical basis in N-dimensional space, and one band establishes a first reference vector in the space. Equal parts of all bands establish a second reference vector. A reference plane contains the first and second reference vectors. The data for the pixel correspond to a point in the space, and that point and the second reference vector define a plane of interest. Hue is an angle between the reference plane and the plane of interest. Chroma is an angle between the point and the second reference vector. Intensity is the point'"'"'s Euclidean norm.

Claims

1. A machine vision method comprising:

sensing light energy associated with a scene, wherein the sensing step comprises measuring three primary color values, wherein the three primary color values are red, green, and blue, wherein the red value is denoted r, the green value denoted g, and the blue value denoted b;

generating color image data representing at least a portion of the scene, wherein the data are arranged as pixels, and the data for a given pixel comprise an intensity value, a hue value, and a chroma value, the intensity value representing the total sensed light energy associated with the pixel, the hue value representing a dominant or average frequency of the light energy associated with the pixel, and the chroma value representing a measure of the light energy on a side of the visible spectrum complementary to the hue, wherein the generating step comprises converting the three primary color values for the given pixel to a set of values comprising the chroma value, the hue value, and the intensity value, and the converting step comprises determining the intensity value (I) in at least approximate accordance with the relation;

I=√

{square root over (r²+g²+b²)}; and

rendering said color image data, wherein the converting step further comprises;

determining a quantity (β

) in at least approximate accordance with the relation;

$\mathrm{\beta <br><br>}={\cos }^{-1}<br><br>\left(\frac{2<br><br>r-g-b}{2<br><br>\sqrt{I^2-\mathrm{rg}-\mathrm{rb}-\mathrm{bg}}}\right);<br><br>\mathrm{and}$

determining the hue value (H) in at least approximate accordance with the relation;

$H=\{\begin{array}{cc}\mathrm{\beta <br><br>},& g><br><br>b\\ 2<br><br>\mathrm{\pi <br><br>}-\mathrm{\beta <br><br>},& \mathrm{otherwise}.\end{array}$

2. The method of claim 1, wherein the light energy associated with the scene comprises reflections of light from an illumination source at an illumination level, whereby the hue value is substantially invariant to changes in the illumination level.

3. The method of claim 1, wherein the light energy associated with the scene comprises reflections of light from an illumination source having an illumination color, whereby the hue value is, but for a correction factor, the same as under white illumination.

4. The method of claim 1, whereby the hue and intensity values are substantially invariant to changes of each other.

5. The method of claim 1, wherein one or more of the determining steps is accomplished by table look-up.

6. The method of claim 1, wherein the converting step further comprises:

determining the chroma value (C) in at least approximate accordance with the relation;

$C={\cos }^{-1}<br><br>\left(\frac{r+g+b}{\sqrt{3}·<br><br>I}\right).$

7. The method of claim 6, wherein one or more of the determining steps is accomplished by table look-up.

8. The method of claim 1 further comprising:

processing the generated color image data.

9. The method of claim 8 further comprising:

after the processing step, converting the chroma, hue, and intensity values to red, green, and blue values.

10. The method of claim 9, wherein the converting step comprises:

determining quantities β

, σ

, and η

in at least approximate accordance with the relations;

$\mathrm{\beta <br><br>}=\{\begin{array}{cc}H,& H\le <br><br>\mathrm{\pi <br><br>}\\ 2<br><br>\mathrm{\pi <br><br>}-H,& \mathrm{otherwise},\end{array}<br><br><br><br>\mathrm{\sigma <br><br>}=\cos <br><br>\left(C\right)·<br><br>\sqrt{3}·<br><br>I,\mathrm{and}<br><br><br><br>\mathrm{\eta <br><br>}=\sqrt{6}·<br><br>I·<br><br>\cos <br><br>\left(\mathrm{\beta <br><br>}\right)·<br><br>\sin <br><br>\left(C\right);<br><br>$

determining the red value (r) in at least approximate accordance with the relation;

$r=\frac{\mathrm{\sigma <br><br>}+\mathrm{\eta <br><br>}}{3};<br><br>$

determining the blue value (b) in at least approximate accordance with the relation;

$b=\frac{1}{2}·<br><br>\left[\left(\mathrm{\sigma <br><br>}-r\right)\pm <br><br>\sqrt{2·<br><br>I^2-{\left(\mathrm{\sigma <br><br>}-r\right)}^2-2·<br><br>r^2}\right];<br><br>\mathrm{and}$

determining the green value (q) in at least approximate accordance with one or more of the relations;

$g=\mathrm{\sigma <br><br>}-r-b=\frac{1}{2}·<br><br>\left[\left(\mathrm{\sigma <br><br>}-r\right)\mp <br><br>\sqrt{2·<br><br>I^2-{\left(\mathrm{\sigma <br><br>}-r\right)}^2-2·<br><br>r^2}\right].$

11. The method of claim 10 further comprising:

determining whether to add or subtract the square root in the relation defining the blue value and the green value, based on the hue value.

12. The method of claim 10, wherein one or more of the determining steps is accomplished by table look-up.

13. The method of claim 10 further comprising:

determining whether the chroma value is legal; and

based on the determining step, conditionally performing the converting step.

14. A method comprising:

sensing electromagnetic energy associated with a source over an area in N frequency bands, wherein N>

1;

generating color image data representing at least a portion of the area, wherein the data are arranged as pixels, and the color data for a given pixel comprise an intensity value, a hue value, and a chroma value, wherein the N frequency bands constitute a mathematical basis in N-dimensional space;

wherein one of the N frequency bands establishes a first reference vector in the space, and equal parts of all N frequency bands establish a second reference vector in the space;

wherein a plane containing the first reference vector and the second reference vector establish a reference plane in the space;

wherein the data for the given pixel corresponds to a point in the N-dimensional space, such that the point corresponding to the data for the given pixel and the second reference vector define a plane of interest, such that the hue value is an angle between the reference plane and the plane of interest, and the chroma value is an angle subtended in the plane of interest between the point corresponding to the data for the given pixel and the second reference vector, and the intensity value is a Euclidean norm of the point corresponding to the data for the given pixel in the space; and

rendering an image based on said color image data.

15. The method of claim 14, wherein N=3.

16. The method of claim 15, wherein the three frequency bands lie at least substantially in the visible spectrum and correspond to the primary colors red, green, and blue.

17. The method of claim 16, wherein the red frequency band establishes the first reference vector.

18. The method of claim 15 further comprising:

processing the generated color image data.

19. The method of claim 18, wherein the processing step comprises:

compressing the generated color image data.

20. The method of claim 14, wherein the source produces electromagnetic energy by reflection of electromagnetic radiation from a separate radiation source, the radiation source having a radiation level, whereby the hue value is substantially invariant to changes in the radiation level.

21. The method of claim 14, wherein the source produces electromagnetic energy by reflection of electromagnetic radiation from a separate radiation source having a dominant radiation frequency, whereby the hue value is, but for a correction factor, the same as under spectrally dispersed radiation.

22. The method of claim 14, whereby the hue and intensity values are substantially invariant to changes of each other.

23. The method of claim 14, further comprising:

performing the generating step for substantially all pixels constituting the color image data.

24. The method of claim 14, further comprising:

performing the generating step for substantially all pixels constituting a contiguous subset of the color image data.

25. The method of claim 14, wherein the generating step comprises:

computing a quantity (β

) according to the formula;

$\mathrm{\beta <br><br>}={\cos }^{-1}<br><br>\left(\frac{N_R·<br><br>N_C}{<br><br>N_R<br><br><br><br><br><br>N_C<br><br>}\right),$

where N_R is a normal vector to the reference plane, and N_C is a normal vector to the plane of interest;

determining the hue value (H) according to the relation;

$H=\{\begin{array}{cc}\mathrm{\beta <br><br>},& \mathrm{if}<br><br><br><br>\mathrm{mid}<br><br>\text{-}<br><br>\mathrm{frequency}<br><br><br><br>\mathrm{energy}><br><br>\mathrm{high}<br><br>\text{-}<br><br>\mathrm{frequency}<br><br><br><br>\mathrm{energy}\\ 2<br><br>\mathrm{\pi <br><br>}-\mathrm{\beta <br><br>},& \mathrm{otherwise};<br><br>\end{array}$

determining the intensity value (I) according to the formula;

$I=\sqrt{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br><br><br>E_{\mathrm{fj}}^2},$

wherein the electromagnetic energy sensed in the N frequency bands are denoted E_f1, E_f2, . . . , E_fN, respectively; and

determining the chroma value (C) according to the formula;

$C={\cos }^{-1}<br><br>\left(\frac{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>E_{\mathrm{fj}}}{\sqrt{N}<br><br>I}\right).$

26. The method of claim 25, wherein N=3.

27. The method of claim 26, wherein the three frequency bands correspond at least approximately to visible red, green, and blue.

28. A color image data processing method comprising:

obtaining color image data representing at least a portion of the area from which electromagnetic energy emanates in at least N frequency bands, wherein the data are arranged as pixels, and the color data for a given pixel comprise an intensity value, a hue value, and a chroma value, wherein the frequency bands constitute a mathematical basis in N-dimensional space;

wherein one of the N frequency bands establishes a first reference vector in the space, and equal parts of all N frequency bands establish a second reference vector in the space;

wherein a plane containing the first reference vector and the second reference vector establish a reference plane in the space;

wherein the data for the given pixel corresponds to a point in the N-dimensional space, such that the point corresponding to the data for the given pixel and the second reference vector define a plane of interest, such that the hue value is an angle between the reference plane and the plane of interest, and the chroma value is an angle subtended in the plane of interest between the point corresponding to the data for the given pixel and the second reference vector, and the intensity value is a Euclidean norm of the point corresponding to the data for the given pixel in the space;

processing one or more of the chroma, hue, and intensity values; and

rendering an image based on said color image data.

29. The method of claim 28 wherein the processing step comprises:

compressing one or more of the chroma, hue, and intensity values, whereby the one or more compressed values can be stored, transmitted, or further processed more efficiently.

30. The method of claim 29 wherein the compressing step comprises:

quantizing the intensity value; and

quantizing the chroma and hue values.

31. The method of claim 30 wherein the two quantizing steps are performed independently of each other.

32. The method of claim 30 wherein the two quantizing steps are performed in parallel.

33. The method of claim 30 wherein the step of quantizing the intensity value results in a quantized intensity value having nine bits or less.

34. The method of claim 30 wherein the step of quantizing the chroma and hue values results in quantized chroma and hue values having a total of nine bits or less.

35. The method of claim 29 further comprising:

transmitting the compressed one or more of the chroma, hue, and intensity values.

36. The method of claim 29, further comprising:

storing the compressed one or more of the chroma, hue, and intensity values.

37. The method of claim 29, further comprising:

processing the compressed one or more of the chroma, hue, and intensity values.

38. The method of claim 28, wherein the obtaining step comprises:

sensing electromagnetic energy associated with a source over an area; and

generating the color image data on the basis of the results of the sensing step.

39. The method of claim 28, wherein the obtaining step comprises:

receiving a transmission comprising the color image data.

40. The method of claim 28, wherein the obtaining step comprises:

retrieving the color image data from a storage memory.

41. The method of claim 40, wherein N>

3, the color image data represents a hyperspectral image, and the rendered color image data in the RGB format is a false color image.

42. The method of claim 28, wherein the processing step comprises converting the color image data to an RGB format, and the method further comprises:

rendering the color image data in the RGB format on a display.

43. A computer-readable medium on which is embedded computer software, the software performing a method for generating color image date representing at least a portion of an area from which electromagnetic energy emanates, the method comprising:

generating color image data representing at least a portion of the area, wherein the data are arranged as pixels, and the color data for a given pixel comprise an intensity value, a hue value, and a chroma value, wherein the N frequency bands constitute a mathematical basis in N-dimensional space;

wherein one of the N frequency bands establishes a first reference vector in the space, and equal parts of all N frequency bands establish a second reference vector in the space;

wherein a plane containing the first reference vector and the second reference vector establish a reference plane in the space;

wherein the data for the given pixel corresponds to a point in the N-dimensional space, such that the point corresponding to the data for the given pixel and the second reference vector define a plane of interest, such that the hue value is an angle between the reference plane and the plane of interest, and the chroma value is an angle subtended in the plane of interest between the point corresponding to the data for the given pixel and the second reference vector, and the intensity value is a Euclidean norm of the point corresponding to the data for the given pixel in the space; and

rendering an image based on said color image data.

44. A machine vision method comprising:

sensing light energy associated with a scene;

generating color image data representing at least a portion of the scene, wherein the data are arranged as pixels, and the data for a given pixel comprise an intensity value, a hue value, and a chroma value, the intensity value representing the total sensed light energy associated with the pixel, the hue value representing a dominant or average frequency of the light energy associated with the pixel, and the chroma value representing a measure of the light energy on a side of the visible spectrum complementary to the hue

processing the generated color image data;

after the processing step, converting the chroma, hue, and intensity values to red, green, and blue values, wherein the converting step comprises;

determining quantities β

, σ

, and η

in at least approximate accordance with the relations;

$\mathrm{\beta <br><br>}=\{\begin{array}{cc}H,& H\le <br><br>\mathrm{\pi <br><br>}\\ 2<br><br>\mathrm{\pi <br><br>}-H,& \mathrm{otherwise},\end{array}<br><br><br><br>\mathrm{\sigma <br><br>}=\cos <br><br>\left(C\right)·<br><br>\sqrt{3}·<br><br>I,\mathrm{and}<br><br><br><br>\mathrm{\eta <br><br>}=\sqrt{6}·<br><br>I·<br><br>\cos <br><br>\left(\mathrm{\beta <br><br>}\right)·<br><br>\sin <br><br>\left(C\right);<br><br>$

determining the red value (r) in at least approximate accordance with the relation;

$r=\frac{\mathrm{\sigma <br><br>}+\mathrm{\eta <br><br>}}{3};<br><br>$

determining the blue value (b) in at least approximate accordance with the relation;

$b=\frac{1}{2}·<br><br>\left[\left(\mathrm{\sigma <br><br>}-r\right)\pm <br><br>\sqrt{2·<br><br>I^2-{\left(\mathrm{\sigma <br><br>}-r\right)}^2-2·<br><br>r^2}\right];<br><br>\mathrm{and}$

determining the green value (g) in at least approximate accordance with one or more of the relations;

$g=\mathrm{\sigma <br><br>}-r-b=\frac{1}{2}·<br><br>\left[\left(\mathrm{\sigma <br><br>}-r\right)\mp <br><br>\sqrt{2·<br><br>I^2-{\left(\mathrm{\sigma <br><br>}-r\right)}^2-2·<br><br>r^2}\right];<br><br><br><br>\mathrm{and}$

rendering said color image data.

45. The method of claim 44, wherein the light energy associated with the scene comprises reflections of light from an illumination source at an illumination level, whereby the hue value is substantially invariant to changes in the illumination level.

46. The method of claim 44, wherein the light energy associated with the scene comprises reflections of light from an illumination source having an illumination color, whereby the hue value is, but for a correction factor, the same as under white illumination.

47. The method of claim 44, whereby the hue and intensity values are substantially invariant to changes of each other.

48. The method of claim 44, wherein the sensing step comprises measuring three primary color values;

and the generating step comprises converting the three primary color values for the given pixel to a set of values comprising the chroma value, the hue value, and the intensity value.

49. The method of claim 48, wherein the three primary color values are red, green, and blue.

50. The method of claim 49, wherein the red value is denoted r, the green value denoted g, and the blue value denoted b, and the converting step comprises:

determining the intensity value (I) in at least approximate accordance with the relation;

I=√

{square root over (r²+g²+b²)}.

51. The method of claim 50, wherein the converting step further comprises:

determining a quantity (β

) in at least approximate accordance with the relation;

$\mathrm{\beta <br><br>}={\cos }^{-1}<br><br>\left(\frac{2<br><br>r-g-b}{2<br><br>\sqrt{I^2-\mathrm{rg}-\mathrm{rb}-\mathrm{bg}}}\right);<br><br>\mathrm{and}$

determining the hue value (H) in at least approximate accordance with the relation;

$H=\{\begin{array}{cc}\mathrm{\beta <br><br>},& g><br><br>b\\ 2<br><br>\mathrm{\pi <br><br>}-\mathrm{\beta <br><br>},& \mathrm{otherwise}.\end{array}$

52. The method of claim 51, wherein one or more of the determining steps is accomplished by table look-up.

53. The method of claim 50, wherein the converting step further comprises:

determining the chroma value (C) in at least approximate accordance with the relation;

$C={\cos }^{-1}<br><br>\left(\frac{r+g+b}{\sqrt{3}·<br><br>I}\right).$

54. The method of claim 53, wherein one or more of the determining steps is accomplished by table look-up.

55. The method of claim 44, further comprising:

determining whether to add or subtract the square root in the relation defining the blue value and the green value, based on the hue value.

56. The method of claim 44, wherein one or more of the determining steps is accomplished by table look-up.

57. The method of claim 44, further comprising:

determining whether the chroma value is legal; and

based on the determining step, conditionally performing the converting step.