METHOD AND SYSTEM FOR COLOR LOOK UP TABLE GENERATION

• US 20130077859A1
• Filed: 09/13/2012
• Published: 03/28/2013
• Est. Priority Date: 09/27/2011
• Status: Active Grant

• Current Assignee: Magnolia Licensing LLC (IP Edge LLC)
• Original Assignee: Jurgen Stauder, Vincent Demoulin
• Inventors: DEMOULIN, Vincent, STAUDER, Jrgen, FARIDUL, Hasan SHEIKH
• Granted Patent: US 8,724,893 B2
• Time in Patent Office: Days
• US Class Current: 382/162
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]
• CPC Class Codes: H04N 1/6052 Matching two or more picture signal generators or two or more picture reproducers
• CPC Class Codes: H04N 13/133 Equalising the characteristics of different image components, e.g. their average brightness or colour balance
• CPC Class Codes: H04N 13/15 for colour aspects of image signals

Abstract

In accordance with an exemplary embodiment, a method is provided to generate a color look up table for multiple images. The method includes selecting color points from a first image. The first color points comprise weak image texture. The method also includes identifying feature points in a spatial neighborhood of each first color point. Additionally, the method includes identifying feature points from a second image. The second feature points correspond to the first feature points. The method also includes determining a color points in the second image. The second color points correspond to the first color points, and are determined using a spatial interpolation of the first plurality of color points from the first plurality of feature points and then applying the obtained spatial interpolation to the second plurality of feature points. Further, the method includes generating the color lookup table.

Claims

1. A method for generating a color lookup table for a plurality of images, comprising:

selecting a first plurality of color points comprising weak image texture, from a first image;

identifying a first plurality of feature points that lie in a spatial neighborhood of each of the first plurality of color points;

identifying a second plurality of feature points from a second image, wherein the second plurality of feature points correspond to the first plurality of feature points;

determining a second plurality of color points in the second image corresponding to the first plurality of color points, using a spatial interpolation of the first plurality of color points from the first plurality of feature points and then applying the obtained spatial interpolation to the second plurality of feature points; and

generating the color lookup table, including a plurality of input colors comprising the first plurality of color points, and a plurality of output colors comprising the second plurality of color points.

2. The method according to claim 1, comprising:

identifying a third plurality of feature points from a third image,

wherein the third plurality of feature points correspond to the second plurality of feature points;

determining a third plurality of color points in the third image corresponding to the second plurality of color points, using a spatial interpolation of the second plurality of color points from the second plurality of feature points and then applying the obtained spatial interpolation to the third plurality of feature points; and

generating the color lookup table, including a plurality of input colors comprising the second plurality of color points, and a plurality of output colors comprising the third plurality of color points.

3. The method according to claim 1, wherein selecting the first plurality of color points comprises:

measuring a variance of signal in each pixel of the first image for each of a red color channel, a green color channel, and a blue color channel;

calculating a global variance of the plurality of pixels and color channels; and

selecting a plurality of candidate color points from the plurality of pixels, wherein all of the following are less than the global variance and multiplied by a constant between zero and one;

the variance of signal for the red color channel;

the variance of signal for the green color channel; and

the variance of signal for the blue color channel.

4. The method according to claim 1, comprising selecting the feature points using a Scale Invariant Feature Transform for the first image and the second image.

5. The method according to claim 1, wherein determining the second plurality of color points comprises:

triangulating each color point of the first plurality of color points by associating a triangle to each color point, each triangles having three edge points, the edge points being selected from the first set of feature points and the four corners of the first image in a way that the color point is included in the triangle;

for each color point of the first plurality of color points and its associated triangle, determining the position of a color point in the second image corresponding to the color point of the first plurality of color points by identifying the corresponding triangle in the second image and spatial interpolation of color points from triangle edge points; and

identifying a color of the color point in the second image.

6. The method according to claim 1, wherein determining the second plurality of color points comprises detecting pixels comprising a weak texture in the second image, wherein the second plurality of color points comprises corresponding color points that show weak texture in the first image and the second image.

7. A system for generating a color lookup table for a first image and a second image, comprising:

a processor; and

a tangible, machine readable memory comprising code configured to direct the processor to;

select a first plurality of color points comprising weak image texture, from the first image;

identify a first plurality of feature points that lie in a spatial neighborhood of each of the first plurality of color points;

identify a second plurality of feature points from the second image based on a scale invariant feature transform of the first plurality of feature points, wherein the second plurality of feature points correspond to the first plurality of feature points;

determine a second plurality of color points in the second image corresponding to the first plurality of color points, using a spatial interpolation of the first plurality of color points from the first plurality of feature points and then applying the obtained spatial interpolation to the second plurality of feature points; and

generate the color lookup table, including a plurality of input colors comprising the first plurality of color points, and a plurality of output colors comprising the second plurality of color points.

8. The system according to claim 7, wherein the tangible, machine readable memory comprises code configured to direct the processor to:

identify a third plurality of feature points from a third image, wherein the third plurality of feature points correspond to the second plurality of feature points;

determine a third plurality of color points in the third image corresponding to the second plurality of color points, using spatial interpolation from the second plurality of feature points to the third plurality of feature points; and

generate the color lookup table, including a plurality of input colors comprising the second plurality of color points, and a plurality of output colors comprising the third plurality of color points.

9. The system according to claim 7, wherein selecting the first plurality of color points comprises:

measuring a variance of signal in each pixel of the first image for each of a red color channel, a green color channel, and a blue color channel;

calculating a global variance of the plurality of pixels; and

selecting a plurality of candidate color points from the plurality of pixels, wherein all of the following are less than the global variance and multiplied by a constant between zero and one;

the variance of signal for the red color channel;

the variance of signal for the green color channel; and

the variance of signal for the blue color channel.

10. The system according to claim 7, wherein the tangible, machine readable memory comprises code configured to direct the processor to select the feature points using a Scale Invariant Feature Transform for the first image and the second image.

11. The system according to claim 7, wherein determining the second plurality of color points comprises:

triangulating each of the first plurality of color points;

identifying a triangle in the second image corresponding to a triangle in the first image comprising a color point of the first plurality of color points;

determining a position of a color point in the second image corresponding to the color point of the first plurality of color points; and

identifying a color of the color point in the second image.

12. The system according to claim 7, wherein determining the second plurality of color points comprises detecting pixels comprising a weak texture in the second image, wherein the second plurality of color points comprises corresponding color points that show weak texture in the first image and the second image.