WO2008088318A1

WO2008088318A1 - Color correction involving color phase detection and phase-dependent control

Info

Publication number: WO2008088318A1
Application number: PCT/US2007/000293
Authority: WO
Inventors: Youichi Nakasone; Toshihiro Sasai
Original assignee: Nucore Technology Inc.
Priority date: 2007-01-05
Filing date: 2007-01-07
Publication date: 2008-07-24
Also published as: WO2008088318A9

Abstract

Color correction is performed on a f first set of three pixel values by determining a color phase of the pixel values. The determined color phase is used to determine a phase difference, and the phase difference is used to control an amount of color phase rotation applied to the chrominance pixel values of the first set . The color phase is also used to determine a first gain, and the first gain is used to control a scaling of the rotated chrominance pixel values, thereby generating color corrected chrominance pixel values. The color phase is also used to determine a second gain, and the second gain is used, to control an amount of scaling applied to the luminance pixel value of the first set, thereby generating the color corrected luminance pixel value. How color phase determines phase difference, the first gain and the second gain is changed depending on lighting conditions .

Description

COLOR CORRECTION INVOLVING COLOR PHASE DETECTION AND PHASE - DEPENDENT CONTROL

Youichi Nakasone Toshihiro Sasai

TECHNICAL FIELD [0001] The present invention relates to color correction

BACKGROUND

[0002] In a typical digital image capture device (for example, a digital camera or cellular telephone having digital camera functionality), an image sensor captures an image as a matrix of pixel values . The image sensor

(for example, a CCD image sensor or a CMOS image sensor) may output the values as digital values or may communicate information to another integrated circuit

(sometimes called an Analog Front End/Timing Generator or "AFE/TG" ) that in turn outputs digital values. The digital values are often in Bayer format. The Bayer format values are often converted into a set of three tristimulus values (such as a set of three RGB pixel values) . For each pixel, there is one such R (red) pixel value, one such G (green) pixel value, and one such B

(blue) pixel value.

[0003] The image sensor typically uses a different color filter for each color detected. For example, a first color filter is disposed over a sensor for red, a second color filter is disposed over a sensor for green, and a third color filter is disposed over a sensor for blue. The operation of the image sensor and color filters is such if light of a pure red color is incident on the image sensor, then the image sensor outputs RGB pixel values chat involve not only a ted pixel value, but also involve a non-7.Pio grpen pixel value and/or noπ-?ero blue pixel value. The RGB pixel values oucpuc by the imaqe sensor are therefore not: pixel values for pure red even though pure red light was incident on the image sensor. The presence of other color components is said to be a "color error." Similarly, it light ot a pure green color is incident upon the image sensor, then the image sensor may output RGB pixel values that involve not only a green pixel value, but also involve a non-zero red pixel value and/or a non-zero blue pixel value.

[0004] So called "color correction" is therefore performed on each set of RGB values as output from the image sensor in order to convert the RGB sensor output values into "corrected" RGB values. Figure 1 (Prior Art) is a diagram that illustrates a set of three unconverted RGB pixels values in a column vector at the right . The "R" represents a red pixel value. The "G" represents a green pixel value. The "B" represents a blue pixel value. The column vector is multiplied by first 3x3 conversion matrix using matrix mathematics to generate a corresponding' set of three converted R' G' B' pixel values in a column vector at the left. The "R'", "G'" and "B'" represent the red, green and blue pixel values as converted (i.e., corrected) .

[0005] Figure 2 (Prior Art) illustrates an example where pure reel light is detected as a set of three RGB pixel values of (100,50,50). The "100" in this notation indicates an intensity of red. The next "50" indicates an intensity of green. The last "50" indicates an intensity of blue. Note that there is significant color error in that there are significant components of green and blue in addition to red. The first 3x3 matrix is applied such that a set of three R' G' B' pixel values of (200, 0, 0) is output. Note that the correction works properly : t; L haL cheie is no giee;. ■_.: blue component in the resulting converted pixel values (200.0,0) .

[0006] Figure 3 (Prior Art) illustrates how three different sets o£ RGB pixels values are converted using the first matrix. The "ftl" above the arrow represents use of the first matrix. The uppermost conversion is the conversion illustrated above in Figure 2 for a condition where pure red light is incident upon the image sensor. The next lower conversion is a conversion for a condition where pure green light in incident upon the image sensor. The resulting pixel values (0,200,0) is corrected in that they involve no red. component or blue component . The bottom-most conversion is a conversion for a condition where pure blue light is incident upon the sensor. The resulting pixel values (0,0,200) is corrected, in the that the values involve no red. or green component .

[0007] Figure 4 (Prior Art) illustrates two transformations. The upper transformation illustrates the transformation of pixel values (75,50,75) output by the image sensor when pure magenta light is incident upon the image sensor. The red. pixel value and the blue pixel value are both 75. The value is therefore said, to be "balanced." . When the first matrix is applied, the result

(100,0,100) properly involves equal intensities of the red and blue components, and no component of green. A typical image sensor may, however, not necessarily output balanced pixel values if pure magenta light is incident upon the image sensor. The image sensor may, for example, output an unbalanced value of (75,50,70) . The red and blue components in the unbalanced value are not identical .

[0008] The lower transformation illustrated in Figure 4 illustrates the result (105,5,85) when the first matrix is applied in an attempt to color-correct the unbalanced

(75,50,70) pixel values. The (105,5,85) result is ..ncorrcct :n thai che intensities of red and blue differ, Ci πϋ a 1 s< ! in thar there is a non-zero amount of green

[0009] Figure 5 illustrates a second matrix The unbalanced (75,50,70) pixel values are multiplied by the second ma t i i x to output, a "corrected" value (100,0,100) . The result is correct in that the red and blue components are equal, and there is no green component.

[OOIOJ To perform a color correction operation on an image having many sets of RGB pixel values, each set of three RGB pixel values is treated separately. If the RGB pixel value is one of the sensor output values that would have resulted were pure red, pure green, or pure blue to have been incident on the image sensor as indicated in Figure 3, then the first matrix is applied to perform color correction on the RGB pixel values. If, on the other hand, the set of RGB pixel values is the set of unbalanced sensor output values that would have resulted if magenta were incident upon the image sensor as indicated in the bottom example of Figure 4, then the second matrix is applied. A decision is therefore made, on a pixel-by-pixel basis, as to which matrix is to be used, to perform the color correction.

[0011] Figure 6 illustrates three additional transformations. The upper transformation illustrates a transformation of balanced RGB pixel values (75,75,50) that would be output from the image sensor if pure yellow light were incident upon the image sensor. The RGB pixel values are "balanced" because the intensities of red and green are identical. If the first matrix is applied, then a proper RGB value of (100,100,0) is output as indicated by the uppermost transformation. Again, an image sensor may not output a balanced value when pure yellow is incident upon it. The image sensor may, for example, output an RGB value of (75,70,50) . As the second transformation of Figure 6 illustrates, applying the f :rs: τ,a:r :x to such an unbalanced value results in incuiieL-L RGB values !10S,85,5) . The r<=^»ri and blue components are not equal, and there is a small amount of blue As the third transformation of Figure 6 j-lluBLiat.es, applying the second matrix also does not result in correct RGB values. The output is (80,135,20) . The red and green components are not equal, and there is an amount of blue.

[0012] Figure 7 (Prior Art) illustrates a third matrix that properly converts the unbalanced yellow sensor pixel values (75,70,50) of Figure 6 into a converted RGB pixel values (100,100,0) . The red. and green components are equal, and there is no blue component. Accordingly, for each set of RGB pixel values, a determination is made as to which one of the three matrices (matrix 1, matrix 2_f or matrix 3) will be used. Using the three matrices of these examples, however, it is not possible to perform several transformations that might be necessary to perform accurate color correction. Figure 8 (Prior Art) illustrates three such transformations that cannot be performed using the three exemplary matrices. A significant number of matrices therefore may be employed to perform adequately accurate color correction in a digital image capture device.

[0013] Figure 9 (Prior Art) is a diagram that illustrates a use of six different matrices to perform so called

"color correction" in one prior art digital camera. The two dimensional diagram illustrates the YCbCr color space . Each pixel involving a set of three RGB pixel values can be converted using a well known conversion matrix into another set of three YCbCr pixel values. The YCbCr- pixel values are said to be in the YCbCr "color space" whereas the RGB pixel values are said to be in the "RGB" color space. Color and luminance information about a pixel can be represented by a set of three RGB pixel values and ca;i a.' si be represented by a set of thr^e YCbCr pixel values In the YCbCr format, the Y represents brightness (or luminance) of the pixel. The Cb and Ct values define t. he color (or chrominance) of the pixel . The two dimensional diagram of Figure 9 therefore represents a graph of all possible chrominances that a pixel can have. The color space of the diagram is sectioned into six areas 1-6. The boundary between areas 1 and 2, for example, is defined by the blue and red pixel values of a pixel being identical (B-R=O) . The boundary between areas 1 and 6, for example, is defined by the green and blue pixel values of a pixel being identical (G-B=O) . Pixels falling within area 1 have values where R-G>0, G-B<0 and B-R<0. In this way, for a given set of RGB pixel values, a determination is made as to which area the RGB pixel value belongs . A different matrix is applied for each area. The color space sectioning of Figure 9 is performed as a way to determine which one of six matrices should be applied to achieve the best color correction.

[0014] Consider a situation in which a first set of three uncorrected RGB pixel values is disposed at location 7 close to the boundary where B-R=O. B-R for the set of RGB pixels is positive, but only slightly so. The set of uncorrected RGB pixels is determined to be in area 1, so a first matrix is applied to perform color correction. Next, consider a situation in which a second set of three uncorrected RGB pixel values is disposed at location 8 close to the boundary where B-R=O. B-R for the set of RGB pixels is negative, but only slightly so. The set of uncorrected RGB pixels is determined to be in area 2, so a second matrix is applied to perform color correction. Even though there is only a slight difference between the values in the first and second sets of uncorrected RGB pixel values, entirely different matrices are applied to the iwo sets c t pixe. values The result oϊ the color correction opeidlion α^.iuj tht- scheme of Finut e 9 is ΛH undesirable ] ump or dispaπcy in the color corrected RGB output values when c he uncorrected RGB input values only exhibit a slight chaiuje, one with respect to another A solution is desired.

SUMMARY

[0015] Color correction is performed on a first set of three pixel values by determining a color phase of the pixel values. In one example, the first set of pixel values (Y¹, Cb¹, Cr¹) is in the YCbCr color space. The color phase is determined from the Cb¹ and Cr¹ chrominance values of the pixel . The determined color phase is then used to determine a phase difference. The phase difference is used to control an amount of color phase rotation applied to the chrominance pixel values of the first set . How the color phase determines the phase difference is a function, and this function is chosen to perform the correct amount of color rotation at each color phase.

[0016] The determined color phase is also used to determine a first gain. The first gain is used to control a scaling of the rotated chrominance pixel values, thereby generating color-corrected chrominance pixel values Cb² and Cr². How the color phase determines the first gain is a function, and this function is chosen to perform the correct amount of scaling at each color phase .

[0017] The determined color phase is also used to determine a second gain. The second gain is used to control an amount of scaling applied to the Y¹ luminance pixel value of the first set, thereby generating the color-corrected luminance pixel value Y². How color phase determines the second gain is chosen to perform the correct amount, ct s_al :^rg a ~ ea . ^v. color phase The col corrected pixel JΛIUP

a •^■ Pn 'c ⁽Y'^', Cb², Cr²)

[0018] In one embodiment, me functions chat determine how the color phase determines the phase difference, how the color phase determines the first gain, and how the color phase determines the second gain are implemented in lookup table memories. An image capture device (for example, a digital camera or a cellular telephone having digital camera functionality) implements the color correction described above. The image capture device has a plurality of light condition settings. Different lookup table values are used depending on the lighting condition setting in which the image capture device is operating .

[0019] Other embodiments and advantages are described in the detailed description below. This summary does not purport to define the invention. The invention is defined by the claims .

BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, where like numerals indicate like components, illustrate embodiments of the invention.

[0021] Figure 1 (Prior Art) is a diagram that illustrates conventional color correction wherein a conversion matrix is used to convert a set of pixel values into a color- corrected set of pixel values .

[0022] Figure 2 (Prior Art) illustrates an example where pure red light is detected as a set of three RGB pixel values of (100,50,50) and these pixel values are color corrected using the matrix of Figure 1.

[0023] Figure 3 (Prior Art) illustrates three examples of color correction using the matrix of Figure 1.

[0024] Figure 4 (Prior Art) illustrates two examples of how pixel values obtained by sensing pure magenta light on r wo different: image sensors are convert, ed using the Tiarrix of Figure 1

[0025] Figure 5 (Pr JO: Art! illustrates a second matrix usable co color correct, the unbalanced pixel values of Figure 4.

[002S] Figure 6 (Prior Art) illustrates three examples of how pixel values obtained by sensing pure yellow light on two different image sensors might be converted using the matrices of Figures 2 and 5.

[0027] Figure 7 (Prior Art) illustrates how a third matrix can be used Co color correct a sec of unbalanced pixel values of Figure 6.

[0028] Figure 8 (Prior Art) illustrates three color correction conversions chat cannot be performed using the three matrices of Figures 2 , 5 and. 7.

[0029] Figure 9 (Prior Art) illustrates how a set of pixel values is color-corrected using a selected one of six different matrices. The matrix selected to correct a set of pixel values depends on which one of six areas of the CbCr color space contains the set of pixel values _

[0030] Figure 10 is a simplified block diagram of a novel image capture device that performs a novel color correction method.

[0031] Figure 11 is a diagram that illustrates how a phase angle is determined from the first Cb¹ and Cr¹ chrominance pixel values.

[0032] Figure 12 illustrates a function for using a phase angle value to determine a phase difference value.

[0033] Figure 13 illustrates how a phase difference value is usable to perform color rotation, thereby converting the first Cb¹ and Cr¹ chrominance values into intermediate Cb¹ and Cr¹ chrominance values.

[0034] Figure 14 illustrates a function for using a phase angle value to determine a first gain value (S_GAIN) . [0035] Fiyuie 15 i 11 ust i a ' (- -= r ^ a ^{f •} =;t gain value iS_GΛIM is usable co perform chtorr.p.ance scaling (gain adjustment) on the i ntermediar e cn.romir.ance values Cb¹ and *" r

[0036] Figure 16 illustrates a function for using a phase angle value to determine a second gain value (B_GAIN) [0037] Figure 17 illustrates how a second gain value (B_GA1N) is usable to perform luminance scaling (gain adjustment) on the first Y¹ luminance value, thereby generating a second, luminance value Y^J . The pixel values Y², Cb² and Cr² are the color-corrected pixel values generated by the novel image capture device of Figure 10. [0038] Figure 18 is a diagram of another embodiment. Color correction in the embodiment of Figure 18 is performed in the HSB (Hue, Saturation, Brightness) color space .

DETAILED DESCRIPTION

[0039J Reference will now be made m detail to some embodiments of the invention, examples of which are illustrated in the accompanying drawings.

[0040] Figure 10 is a diagram of an image capture device 10 in accordance with one novel aspect. Image capture device 10 may, for example, be a digital camera or a mobile communication device that includes digital camera functionality. Image capture device 10 includes a sensor portion 11 and a color correction portion 12. Color correction portion 12 receives pixel information from the sensor portion and performs color correction on the pixel information by determining a color phase of the pixel information, and then by using the color phase to control a color phase rotation operation, a chrominance scaling operation, and a luminance scaling operation. In the illustrated example, sensor portion 11 includes an image sensor and analog front end/timing generator (AFE-TG) 13, a Bayer -to RGB conversion circuit 14 and an RGB co YCbCr i.'j.'iveisiυπ Lircuit 15

[0041] lhe image sensor and AFE/TG circuitry captures an image and outputs corresponding Bayer data Bayer- to- RGB conversion circuit 14 converts the Bayer data into sets of tristimulus RGB pixel values One set of RGB pixel values includes a red (R) pixel value, a green (G) pixel value, and a blue (B) pixel value. There is one such set of RGB pixel values for each pixel.

[0042] Operation of the xcnage capture device 10 of Figure 10 is described in connection with one such set of RGB pixel values being supplied to the RGB- to- YCbCr conversion circuit 15. The RGB-to-YCbCr conversion circuit 15 converts the set of RGB pixel values into a first set 26 of pixel values in the YCbCr color space. This first set 26 of pixel values involves a first Y¹ luminance value, a first Cb¹ chrominance value, and a first Cr¹ chrominance value. The first Cb¹ chrominance value and the first Cr¹ chrominance value are supplied to a polarization block 16. Polarization block 16 converts the first Cb¹ chrominance value and the first Cr¹ chrominance value into a corresponding phase angle phi (Φ) .

[0043] Figure 11 is an illustration of the conversion operation performed by polarization block. 16. The pair of first Cb¹ and Cr¹ chrominance values corresponds to a point in the X-Y plane illustrated in Figure 11. The phase angle phi from the origin is the arctangent of Cr¹ZCb¹. The phase angle phi is represented as a ten bit number on parallel bus 17. Values in the range of from 0 to 1024 represent corresponding values in the range of from zero degrees to 360 degrees.

[0044] The phase angle phi is supplied in parallel to a color phase adjust circuit 18, a chrominance adjust circuit 19, and a luminance adjust circuit 20. A tur.ci ;or. block 21 within color phase ad3 use circuit IB coπver L .•=■ t he phase angle phi into a rr>r respond 1 r^*:q phase ditference value Λs Φ) .

[0045] Figure 12 is a diagram chat illustrates the cot

between the phase angle phi (Φ) suppl i ed to function block 21 and the phase difference value Δ(Φ) output from function block 21. In the diagram o£ Figure 12, the incoming phase angle phi (Φ) is represented on the X-axis. The resulting phase difference Δ(Φ) is represented on the Y-axis. Depending on the value of the incoming phase angle phi, the output phase difference value ranges between a high value of approximately +25 degrees and a low value of approximately -15 degrees. In the illustrated example, the function that converts the incoming phase angle phi (Φ) into the phase difference value appears as a stepped sinusoidal function. [0046] Returning to Figure 10, the phase difference value Δ(Φ) is supplied to a color phase rotation block 22 of the color phase adjust circuit 18. Color phase rotation block 22 performs a color phase rotation operation that is controlled by the phase difference value Δ(Φ) . Color phase rotation block 22 receives the first Cb¹ chrominance value and the first Cr¹ chrominance value and generates an intermediate Cb¹ chrominance value and an intermediate Cr¹ chrominance value .

[0047] Figure 13 illustrates how an incoming first Cb¹ chrominance value is converted into an intermediate Cb¹ chrominance value depending on the magnitude of the phase difference value. Similarly, the diagram illustrates how an incoming first Cr¹ chrominance value is converted into an intermediate Cr¹ chrominance value depending on the magnitude of the phase difference value. The amount of color phase rotation at each phase angle phi can be preset by adjusting how the function of Figure 12 converts the phase angle phi into the phase difference va

le if I he function ot ^cig i-e ^η.2 generates a phdye di£ [eience value of zero f <_ιi a particular phase angle phi, then there is no color phase rotation performed for the phase angle phi As seen in Figure 12 no color phase rotation is pet formed for phase angles Φ of 0 and 512.

[0048] As illustrated m Figure 10, the phase angle phi Φ is also supplied to the chrominance adjust circuit 19 A gam determination block 23 receives the phase angle phi value and converts it into a corresponding gam value S_GAIN

[0049] Figure 14 illustrates how an incoming phase angle phi is converted into a corresponding S-GAIN value. The incoming phase phi is represented on the X-axis. The resulting S_GAIN value is represented on the Y-axis . The values of the S-GAIN values output from gain determination block 23 range from approximately twenty percent to approximately negative twenty percent. The function that converts the incoming phase angle phi into an S-GAIN value has the appearance of a stepped sinusoidal wave.

[0050] Returning to Figure 10, the S-GAIN value output from gain determination block 23 is supplied to a chrominance gain block 24. The chrominance gain block 24 receives the intermediate Cr¹ chrominance and intermediate Cb¹ chrominance values and scales them in accordance with the value of S-GAIN.

[0051] Figure 15 illustrates how an incoming intermediate Cb¹ chrominance value is scaled to generate an output second Cb² chrominance value depending on the value of S_GAIN. Similarly, the figure illustrates how an incoming intermediate Cr¹ chrominance value is scaled to generate an output second Cr² chrominance value depending on the value of S_GAIN. The second chrominance values are designated in the figure with superscript values of cwi Tl'":>=- SF-ror._'i ~h'^' and Cr' chrominance values ar^e ~.u"p . r from che chiomiαance ad]usL circuit 19 and form :wn values ot a sec ^5 of color corrected pixel values

[0052] As illustrated in Figure 10, t.he phase angle phi is also supplied to the luminance adjust circuit 20 A gain determination, block 26 receives the phase angle phi value and converts it into a corresponding gain value B_GAIN.

[0053] Figure 16 illustrates how an incoming phase angle phi is converted into a corresponding B-GAIN value. The incoming phase phi is represented on the X-axis. The resulting B_GAIN value is represented on the Y-axis. The values of the B-GAIN values output from gain determination block 26 range from approximately positive twenty percent to approximately negative twenty percent . The function that converts the incoming phase angle phi into a B-GAIN value has the appearance of a stepped sinusoidal wave.

[0054] Returning to Figure 10, the B-GAIN value output from gain determination block 26 is supplied to a luminance gain block 27. The luminance gain block 27 receives the first Yl luminance value that is being output by RGB-to-YCbCr conversion circuit 15. Luminance gain block 27 scales the first Y¹ luminance value depending on the B-GAIN value.

[0055] Figure 17 illustrates how the first Y¹ luminance value is scaled to generate a second Y² luminance value depending- on the value of B-GAIN. The second Y² luminance value is designated in the figure with a two superscript . The second Y² luminance value as output from luminance gain block 27 is the Y² luminance value of the second set 25 of color corrected pixel values .

[0056] Although the present invention has been described in connection with certain specific embodiments for instructional purposes, the present invention is not limited Lherec._' I r. i e πedi ace Cb¹ and Cr¹ can be input cc the cht orri :ιan! i- αd]usι ^■. iuuiL 19 and/or to L he 1 urr.

e ad]usc circuit 20 through an additional polarization circuit Tn one embodiment, the Bayer- to-RGB conversion circuit 14, the RGB-to-YCbCr conversion circuit 15 and the color correction portion 12 are all disposed on a single digital image processing integrated circuit. Polarization block 16 may output a number other than a phase angle that is nonetheless indicative of a relationship between the first Cb¹ chrominance value and the fxrst Cr¹ chrominance value. Block 16 may, for example, output a simple ratio of the two first chrominance values. In one embodiment, no block 16 is provided, but rather the two first Cb¹ and Cr¹ chrominance values are supplied directly to lookup blocks 21, 23 and 26. The lookup blocks 21, 23 and 26 use the two first Cb¹ and Cr¹ chrominance values to lookup a phase difference value, an S_GAIN value, and a B_GAIN value, respectively. The chrominance scaling and color phase rotation operations can be performed in either order. Although lookup table (LUT) memories are described above as implementations of blocks 21, 23 and 26, other circuitry for converting one number into another number other than LUT memories can be used. For example, portions of arithmetic logic can perform simple arithmetic operations in order to generate the desired phase difference value, S_GAIN value, and B_GAIN value. In one embodiment, an integrated circuit embodying the color correction circuitry described above has an interface for receiving image data from one or more image sensors that do not output Bayer format data, but rather output image data in RGB format or in another color space format . The interface on the integrated circuit is configurable to receive image data from a selectable one of these different image sensors. [0057] Tn OtIf=- F-rroo>i : -neii r a user of a digiral camera ran selecc one of a plural it¹,' ^f 1 iqht condition seuinqs Alternatively, the camera can put itself into one ot the light condition sett inys For each different light condition setting, the function of phase angle implemented by block ?1 is different A different lookup table memory may, for example, be consulted depending on the light condition setting A single SRAM (static random access memory) lookup table memory may be loaded with different data depending on the light condition setting such that a single lookup table memory can be used for block 21. Also, the function of phase angle implemented by blocks 23 and/or 26 can also be made to be different depending on the light condition setting of the camera. There may, for example, be three or more such light condition settings The spectrum characteristics of a sensor (CCD or. CMOS) may vary depending on the manufacturer of the sensor. In one advantageous aspect, the SRAM lookup table memories are loaded with different data depending on the type of sensor used (for example, CCD or CMOS) in order to compensate for differences between these types of sensors so that any one of multiple different sensors can be used in conjunction, with the same type of color correction integrated circuit in a digital camera. Accordingly, various modifications, adaptations, and combinations of various features of the described embodiments can be practiced without departing from the scope of the invention as set forth in the claims .

Claims

^■"■ [ .A : MΓ

What is claimed is

1 An integrated circuit r.hac processes a first set ot YCbCr pixel values, the first set comprising a first Y luminance value, a first Cb chrominance value, and a first Cr chrominance value, the integrated circuit generating a second set of YCbCr pixel values, the second set comprising a second Y lumuninance value, a second Cb chromxnance value, and a second Cr chrominance value, the integrated circuit comprising: a color phase adjust circuit that receives the first Cb chrominance value and the first Cr chrominance value and that performs color phase rotation based on a relationship between the first Cb chrominance value and the first Cr chrominance value, the color phase adjust circuit outputting an intermediate Cb chrominance value and an intermediate Cr chrominance value; a chrominance adjust circuit that receives the intermediate Cb chrominance value and the intermediate Cr chrominance value and that performs scaling based on the relationship between the first Cb chrominance value and the first Cr chrominance value, the chrominance adjust circuit outputting the second Cb chrominance value and the second Cr chrominance value; and a luminance adjust circuit that receives the first Y luminance value and that performs scaling based on the relationship between the first Cb chrominance value and the first Cr chrominance value, the luminance adjust circuit outputting the second Y luminance value.

2. The integrated circuit of Claim 1, further comprising: a circuit that receives the first Cb chrominance value and the first Cr chrominance value and that outputs a number indicative of the relationship between the first Ch chrominance value and the tirsr Cr chrominance value, wherein the circuit supplier t_he number Lo the color phase adjust circuit, the chrominance adjust circuit and the luminance adjust circuit

3. The integrated circuit of Claim 2, wherein the color phase adjust circuit comprises: a portion that converts the number into a phase difference value; and a portion that uses the phase difference value to convert the first Cb chrominance value and the first Cr chrominance value into the intermediate Cb chrominance value and the intermediate Cr chrominance value.

4. The integrated circuit of Claim 3, wherein the portion that converts the number into the phase difference value comprises a lookup-table memory.

5. The integrated circuit of Claim 1, further comprising : an RGB- to-YCbCr conversion circuit that receives a set of RGB pixel values and that outputs the first set of YCbCr pixel values.

6. The integrated circuit of Claim 5, further comprising: a Bayer-to-RGB conversion circuit that receives Bayer values and outputs the set of RGB pixel values .

7. The integrated circuit of Claim 6, wherein the integrated circuit is part of an image capture device, the image capture device comprising an image sensor, the image sensor outputting the Bayer values. θ The integrated circuit of claim 7 wherein the image sensor comprises an analog front end/ c inning generator (AFE/TG) device and one of a charge coupled sensor (CCD) device and a CMOS sensor device.

9. The integrated circuit of Claim 2, wherein the chrominance adjust circuit comprises: a portion that converts the number into a gain value ; and a portion that uses the gain value to scale the intermediate Cb chrominance value into the second Cb chrominance value and to scale the intermediate Cr chrominance value into the second Cr chrominance value.

10. The integrated circuit of Claim 9, wherein the portion that converts the number into the gain value comprises a lookup-table memory.

11. The integrated circuit of Claim 1, wherein the relationship between the first Cb chrominance value and the first Cr chrominance value is a phase angle.

12. The integrated, circuit of Claim 1, wherein the relationship between the first Cb chrominance value and the first Cr chrominance value is a ratio of the first Cb chrominance value to the first Cr chrominance value.

13. The integrated circuit of Claim. 1, wherein the integrated circuit is part of an image capture device, the image capture device comprising an image sensor, the image sensor outputting image data that is converted into the first set of YCbCr pixel values .

14. A method, comprising: '.a,- using an image sensor to capture an image and co out pur. image data, wherein the image sensor is pare of an irna-je rapture device, the image capture, device being taken from the group consisting of: a digital camera, a mobile communication device that includes a digital camera functionality;

(b) converting the image data into a set of YCbCr image pixel values, the set of YCbCr image pixel values including a first Y luminance value, a first Cb chrominance value and a first Cr chrominance value;

(c) using the first Cb chrominance value and the first Cr chrominance value to determine a phase angle; and

(d) using the phase angle to control a color adjust rotation operation, the color phase adjust operation converting the first Cb chrominance value and the first Cr chrominance value into an intermediate Cb chrominance value and an intermediate Cr chrominance value.

15. The method of Claim 14, wherein the image capture device is operable in one of at least three light condition settings, wherein when the image capture device is in a first of the light condition settings then the phase angle is used in (d) in a first manner to control the color phase adjust operation, wherein, when the image capture device is in a second of the light condition settings then the phase angle is used in (d) in a second manner to control the color phase adjust operation, and wherein when the image capture device is in a third of the light condition settings then the phase angle is used in (d) in a third manner to control the color phase adjust operation. _ i^, T¹^e method of Claim 14, further corrpt ismα

¹F-'_' M =;:ng the phase angle to ront rm a -hromi nance adjust operation, the chrominance adjust opetauon converting the intermediate Cb chrominance value and the inrerTiPdiate Cr chrominance value into second Cb chrominance value and a second Cr chrominance value.

17 The method of Claim 14, wherein the color adjust rotation operation comprises the steps of- converting the phase angle into a phase difference,- and using the phase difference to adjust a color phase of the first Cb chrominance value and the first Cr chrominance value, the Intermediate Cb chrominance value and the intermediate Cr chrominance value being a result of the color phase adjustment.

18. Trie method of^" Claim 17, wherein the converting of the phase angle into a phase difference involves using the phase angle to look up the phase difference in a lookup table memory.

19. An image capture device, comprising: a sensor portion that captures an image and outputs corresponding pixel information; and means for receiving the pixel information from the sensor portion and for performing- color correction on the pixel information by determining a color phase of the pixel information, and then by using the color phase to control a color phase rotation operation that converts the pixel information received from the sensor portion into phase rotated pixel information.

20. The circuit of Claim 19, wherein the color correction also involves using the color phase to control a c. h r < >nι :. :iα :."÷ ijaiii operation, rhp rhror-. ^riaτ"f^> ya:r, operation '---.-.nvc r t ing Che phase rotated pixel information into roi'-ir i-oti'pcted pixel informar inn.

21 The circuit of Claim 20, wherein the color correction also involves using the color, phase to control a luminance gain operation, the luminance gain operation converting a luminance value of the pixel information received by the means into a luminance value of the color corrected pixel information.

22. The circuit of Claim 19, wherein the sensor portion comprises an image sensor that outputs Bayer data, and wherein the pixel information received by the means is YCbCr pixel information.