WO2005112429A1

WO2005112429A1 - Color correction device, color correction method, and color correction display method

Info

Publication number: WO2005112429A1
Application number: PCT/JP2005/009220
Authority: WO
Inventors: Kenichiro Waki; Takashi Masuda; Kenji Kobayashi; Tsutomu Okuno; Takashi Kurino
Original assignee: Acutelogic Corporation
Priority date: 2004-05-18
Filing date: 2005-05-13
Publication date: 2005-11-24
Also published as: JPWO2005112429A1

Abstract

In a color difference signal space having four axes: Cb+ axis, Cb- axis, Cr+ axis, and Cr- axis, color correction is performed by rotating and extending/contracting the four axes independently from one another. When axes at one side or both sides of the area where a color difference signal (Cb, Cr) to be corrected exists are rotated and extended/contracted, the other axes are also rotated and extended/contracted so as to minimize the affect to the other colors. Thus, when a desired color is subjected to color correction, the affect to the other colors can be reduced and the color continuity at the boundaries of the four axes can easily be maintained, thereby performing a color correction with a high degree of freedom.

Description

Specification

Color correction device, color correction method, and color correction display method

Technical field

The present invention relates to a color correction method and a color correction display method for performing color correction of an image signal.

Front and back technology

2. Description of the Related Art Conventionally, in an imaging device that captures an image of a subject and outputs an image signal, or a display device that displays an input image signal, a color correction device that performs color correction of the image signal in order to improve color reproducibility. This type of color correction device that is used includes a color correction device that performs color correction on an RGB signal that expresses a color using a combination of red (R), green (G), and blue (B) light-primary colors. And color correction for color difference signals separated from RGB signals.

FIG. 19 is a diagram showing the configuration of the former evening color correction device. As shown in Fig. 19, the color correction device of this type has an RGB matrix converter 1

It is configured with 0 0. As shown in Fig. 21, the color correction device in this evening inputs RGB signals (R, G, B) and finds the product of the signals with the 3X3 conversion matrix to obtain the converted values. RGB signal is obtained (R'G ', B'). In this way, by appropriately setting the value of each coefficient C 1 1 C 3 3 of the 3 × 3 conversion matrix and performing the matrix conversion, it is possible to perform color correction on the input RGB signal. . For example, if the skin is too red, redness can be reduced by performing a conversion to reduce the R component of the RGB signal in the RGB signal space. However, when the respective coefficients C 11 to C 33 of the conversion matrix are determined in order to correct a certain color, the conversion matrix affects other colors, and Color correction is performed. As a result, overall color reproduction is often undesirable. Also, when performing color correction on RGB signals, the luminance may change, which is not desirable for color conversion.

FIG. 20 is a block diagram showing the configuration of the latter evening color correction device. As shown in FIG. 20, this type of color correction device includes a luminance / color difference signal separation unit 101 , Chrominance signal matrix conversion unit 102, luminance / chrominance signal synthesis unit 10

3 is configured.

The luminance / chrominance signal separation unit 101 separates the input image signal into a luminance signal (Y) and color difference signals (Cb, Cr) and outputs them.

The chrominance matrix conversion unit 102 receives the chrominance signals (Cb, Cr) separated by the luminance / chrominance signal separation unit 101, and receives the chrominance signals (Cb, Cr). By performing color difference matrix conversion, the converted color difference signal

(C b ′, C r ′) where the color difference signal (C b) is obtained by subtracting the luminance signal (Y) from the B signal, and the color difference signal (C r) is the R signal From the luminance signal (Y).

The luminance and chrominance signal synthesizing unit 103 includes a chrominance signal (Cb ′, Cr ′) after color conversion corrected by the chrominance matrix conversion unit 102 and a luminance and chrominance signal separation unit The image signal is generated by combining the luminance signal (Y) separated in step 1 and output to the outside.

As described above, according to the conventional color correction apparatus shown in FIG. 20, the input image signal is separated into the luminance signal (Y) and the color difference signals (Cb, Cr), and the color difference signals (Cb, Cr) are separated. ) Is subjected to matrix conversion to perform color correction, so that color correction can be performed without changing luminance. However, even in the color correction device of this type, when a specific color is corrected, the other colors are also corrected, which often results in an undesirable color reproduction as a whole. Once. As described above, in the color correction device of the present invention, correction for a specific color has a large effect on other colors in any type, and only a specific color without affecting the reproducibility of other colors. Color correction with some degree of freedom cannot be performed

By the way, in the color difference signal space where the color difference signal (B-Y) is the X-axis and the color difference signal (R-Y) is the Y-axis, the axes C y (X) of Y e (yellow) and B (blue) The hue range is divided into six regions by the axes Mg (magenta) and G (green) of R and R (red), and the input color difference signal (B—YR

) To which area the color difference signal (B—YR

) Has been proposed (see, for example, Patent Document 1). A color correction circuit has been proposed that separates the position vector of) into vector components of two axes that define the region where the hue is different. -1 7 6 6 5 6

This color correction circuit performs a first-order conversion on each of the above-described vector components according to a predetermined conversion matrix, and calculates a color difference based on each of the first-ordered vector components and a coefficient of each of the vector components. The signal (Β—Υ, R—Υ) is converted. According to this conventional technique, color correction can be performed only on a color difference signal within a predetermined area.

However, the technique disclosed in Patent Document 1 has a problem that the degree of freedom of the v. Region division is reduced because the predetermined region is divided by a straight line passing through the origin (the origin is not the starting point). In addition, since color correction is performed only on the color difference signal in a predetermined area, if color correction is performed in a certain area, continuity with colors in other areas is not maintained. In order to maintain color continuity with respect to other areas, it is necessary to adjust the glitter after trial and error in other areas other than the area where color correction is performed. It was very troublesome.

The present invention has been made in order to solve such a problem, and it is possible to reduce the influence on other colors by correcting a desired color, and to improve the color difference signal space. The purpose is to maintain color continuity at the boundaries of the axes. Disclosure of the invention

The color correction device of the present invention receives a separated color difference signal and determines where the color difference signal exists in a plurality of regions divided by the axis of the number of pixels starting from the origin of the color difference signal space and the color difference signal space. Then, the color # 1 is converted using the conversion matrix corresponding to the determined area. The transformation matrix is configured as a matrix having, as elements, a coefficient for independently expanding and contracting a plurality of axes and a rotation angle for independently rotating the plurality of axes.

Further, in another aspect of the present invention, the separated color difference signal is gamma-converted based on a predetermined gamma curve, and the gamma conversion is performed anywhere in a plurality of regions divided by a plurality of axes starting from the origin of the color difference signal space. It is determined whether or not the converted color difference signal exists, and the gamma-converted color signal is converted using a conversion matrix corresponding to the determined area. The transformation matrix is configured as a matrix having rotation angles for rotating a plurality of axes independently as elements.

According to another aspect of the present invention, the color difference signal has a first-order term and a second-order term, and the coefficients are a first-order coefficient for the first-order term and a first-order coefficient for the second-order term. It is characterized by having

According to the present invention, a plurality of axes in the color difference signal space are That is, even if one axis or both axes of the area where the color difference signal for performing color correction exists is rotated and expanded / contracted, the other axes are independently rotated and expanded / contracted so that the influence on other colors is reduced. Since the color can be corrected, it is possible to reduce the influence on other colors when the desired color is corrected. In addition, by rotating a plurality of axes independently, the colors on both sides of the axes are converted following the rotation of the axes, so there is no need to set parameters such as the coefficients of the conversion matrix and the rotation angle. Therefore, the color continuity at the axis boundary can be easily maintained.

, Color correction with a high degree of freedom can be performed, and color reproducibility can be improved.

Further, according to another aspect of the present invention, since the color difference signal is gamma-converted based on a predetermined gamma curve and then converted by a conversion matrix, it is possible to apply a gain to the color difference signal by gamma conversion. it can. At this time, depending on the setting of the gamma curve, the magnitude of the gain can be made variable according to the magnitude of the original color difference signal. As a result, it is possible to apply a non-uniform gain to the color difference in the same area, and it is possible to perform color correction with a higher degree of freedom and to improve color reproducibility. Monkey

Further, according to another aspect of the present invention, since the coefficients used in the transformation matrix have a first-order coefficient and a second-order coefficient, the degree of freedom for setting the coefficients is increased. . As a result, it is possible to apply non-uniform gain to the color difference signals in the same area. Therefore, color correction with higher flexibility can be easily performed, and color reproducibility can be improved.

FIG. 1 is a block diagram illustrating a configuration example of a color correction device according to the first embodiment. FIG. 2 is a flowchart showing the operation and color correction method of the color correction device according to the first embodiment.

FIG. 3 is a diagram illustrating an example of four-axis conversion by the color correction device according to the first embodiment.

FIG. 4 is a diagram illustrating another example of four-axis conversion by the color correction device according to the first embodiment.

FIG. 5 is a diagram illustrating an example of a screen of a thread for calculating a primary coefficient and a rotation angle of a conversion matrix in the color correction device according to the first embodiment.

FIG. 6 is a block diagram illustrating a configuration example of a color correction device according to the second embodiment.

FIG. 7 is a diagram illustrating an example of a graph used for chroma gamma conversion by the color correction device according to the second embodiment.

FIG. 8 is a diagram illustrating an example of four-axis conversion by the color correction device according to the second embodiment.

FIG. 9 is a diagram illustrating another example of a graph used for chroma gamma conversion by the color correction device of the second embodiment.

FIG. 10 is a block diagram illustrating a configuration example of a color correction device according to the third embodiment.

FIG. 11 is a diagram illustrating an example of four-axis conversion by the color correction device according to the third embodiment.

FIG. 12 is a diagram comparing an example of four-axis conversion by the color correction device of the third embodiment with an example of four-axis conversion by the color correction device of the first embodiment. FIG. 13 is a block diagram illustrating a configuration example of a color correction device according to the fourth embodiment.

m 14 is a diagram illustrating an example of 8-axis conversion by the color correction device according to the fourth embodiment. FIG. 15 is a diagram illustrating a modification of the axis alignment performed by the color correction device according to the fourth embodiment.

FIG. 16 is a diagram illustrating an example of a screen of a tool for calculating a primary coefficient, a secondary coefficient, and a rotation angle of a conversion matrix for adding an axis in the color correction apparatus according to the fourth embodiment.

FIG. 17 is a diagram illustrating an example in which axes are freely arranged in the color correction device according to the fourth embodiment.

FIG. 18 is a block diagram showing a configuration example of a color correction device according to the fifth embodiment.

FIG. 19 is a block diagram showing a configuration example of a conventional color correction device. FIG. 20 is a block diagram showing another configuration example of a conventional color correction device. Figure 21 shows the equation that also converts RGB 1s with the conventional color correction m.

It is a figure

FIG. 22 is a diagram showing an equation for converting a color difference signal of the color correction device according to the first embodiment.

23 is a diagram illustrating an expression of a conversion matrix according to the coordinates of the color difference signal of the color correction according to the first embodiment.

24 is a diagram showing an equation for calculating the sum of differences between conversions of a plurality of color difference signals of the color correction device according to the first embodiment.

FIG. 25 is a diagram showing an expression of a conversion matrix when performing gamma conversion of color correction according to the second embodiment.

FIG. 26 is a diagram illustrating a formula for converting a color difference signal including a quadratic term of color correction according to the third embodiment and a conversion matrix formula for coordinates of the color difference signal.

FIG. 27 is a diagram illustrating an equation for converting a color difference signal when performing eight-axis conversion of the color correction device according to the fourth embodiment. FIG. 28 is a diagram showing an equation for converting a color difference signal when the positions of the axes of the color correction device according to the fourth embodiment are freely arranged.

FIG. 29 is a diagram illustrating an example of a matrix of an RGB matrix conversion unit of the color correction device according to the fifth embodiment.

FIG. 30 is a diagram illustrating three matrices for obtaining a first transformation matrix of the color correction device according to the fifth embodiment.

FIG. 31 is a diagram showing a matrix for converting a color difference signal of the color correction device according to the fifth embodiment.

FIG. 32 is a diagram showing an expression for obtaining a first transformation matrix of the color correction device according to the fifth embodiment.

FIG. 33 is a diagram illustrating two matrices for obtaining a second conversion matrix of the color correction device according to the fifth embodiment.

FIG. 34 is a diagram showing an expression for obtaining a second transformation matrix of the color correction device according to the fifth embodiment.

(First Embodiment)

Hereinafter, a first embodiment according to the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration example of a color correction apparatus according to a first embodiment.

. In FIG. 1, reference numeral 1 denotes a luminance / chrominance signal separation unit, which separates an input image signal into a luminance signal (Y) and color difference signals (Cb, Cr) and outputs them.

Reference numeral 2 denotes a four-axis conversion unit (corresponding to the axis conversion unit in the claims), to which the color difference signals (Cb, Cr) separated by the luminance color difference signal separation unit 1 are input. And the Cb component of the color difference signal (cbCr) is the X axis, and the Cr component is

In the color difference signal space on the Y axis, it is determined which quadrant (any one of the first to fourth quadrants) the coordinates of the color difference signal (C b, Cr) are included in. By calculating the product of the conversion matrix C _n (η · is an integer of 1 to 4) and the color difference signal (C b, C r) as shown in FIG. 22, the color difference signal (C b, C r) are color-corrected, and the converted color difference signals (C b ′, C r ′) are obtained.

Reference numeral 3 denotes a luminance / chrominance signal synthesizing unit, which converts the color difference signals (C b ′, Cr ′) after the color correction by the 4-axis conversion unit 2 and the luminance signal (Υ) separated by the luminance / chrominance signal separation unit 1 To generate an image signal and output it to the outside.

The color difference signal space described above is based on the Cb axis for the positive direction of the color difference signal (Cb).

, The Cb axis for the negative direction of the color difference signal (Cb), the Cr axis for the positive direction of the color difference signal (Cr), and the Cr axis for the negative direction of the color difference signal (Cr). The area is divided, and the area between the Cb axis and the Cr axis is in the first quadrant, the area between the Cr and Cb axes is the second quadrant, the Cb-axis and the Cr axis The area between them is the third quadrant, and the area between the Cr single axis and the C b axis is the fourth quadrant

The quadrant that contains the coordinates of the color difference signal (C b, Cr) is determined by the color difference ie

It is determined by the sign of (C b, C r). For example, when the color difference signal (C b) and the color difference signal (C r) are both positive, the color difference signal (C b, C r) exists in the first quadrant. If the color difference signal (C r) is positive (expressed as C b —) and the color difference signal (C r) is positive, the color signal (C b, C r) exists in the second quadrant. Further, both the color difference signal (C b) and the color difference signal (C r) are negative (C b

, C r-), the color difference signals (C b, C r) are in the third quadrant. When the color difference signal (C b) is positive and the color difference signal (C r) is negative (represented as C r 1), the color difference signal (C b, C r) is present in the fourth quadrant. If the coordinates of (C b, C r) exist in the first quadrant, 3 Perform color correction using the transformation matrix C shown in (a). If the coordinates of the color difference signals (C b, C r) exist in the second quadrant, color correction is performed using the conversion matrix C ₂ shown in FIG. 23 (b). If the coordinates of the color difference signal (C b, C r) exist in the third quadrant, color correction is performed using the conversion matrix C ₃ shown in FIG. If the coordinates of r) are present in the fourth quadrant performs color correction by the transformation matrix C ₄ shown in FIG 2 3 (d).

Here, in the equation shown in FIG. 2 3, 0 the rotational angle of the C b axis, 1 primary coefficient indicating the gain of the C b axis (stretching degree) 0 ₂ the rotation angle of the C r axis, 1 ₂ C Primary coefficient indicating r-axis gain, 0 ₃ is C b-axis rotation angle, 1 ₃ f-axis is b-axis gain linear coefficient, 0 ₄ is C r single-axis rotation angle, 1 ₄ f-axis Is a first-order coefficient indicating the gain of. Hue is converted by the rotation angle 0 of each axis, the saturation is converted by the primary coefficient 1 i to 1 ₄ showing the gain of each axis.

Next, an operation of the color correction apparatus according to the first embodiment and a color correction method will be described. FIG. 2 is a flowchart showing an operation of the color correction apparatus according to the first embodiment and a color correction method. First, the luminance signal (Y) and the color difference signals (Cb, Cr) are separated by the luminance / chrominance signal separation section 1 (step S1). The luminance signal (Y) separated by the luminance and chrominance signal separation unit 1 is output to the luminance and chrominance signal synthesis unit 3, and the chrominance signal (C b C r) separated by the luminance and chrominance signal separation unit 4 is Output to axis conversion unit 2.

In the 4-axis conversion unit 2, the luminance and chrominance signal separation unit 1 outputs the chrominance signals (Cb, Cr

) Is input to check which quadrant in the chrominance signal space exists (step S 2). Then, the color difference signals (C b, C r) are converted by using a conversion matrix corresponding to the quadrant (step S 3). The four-axis converter 2 outputs the converted color difference signals (C b ′, Cr −) to the luminance / color difference signal synthesizer 3. The luminance / chrominance signal synthesizing unit 3 synthesizes the luminance signal (Y) input from the luminance / chrominance signal separating unit 1 and the converted color difference signals (C b ′, C r ′) input from the 4-axis conversion unit 2. To generate an image signal and output it to the outside (Step S4

Next, a specific example of the 4-axis conversion of the color correction method according to the first embodiment will be described with reference to the drawings. FIG. 3 is a diagram illustrating an example of four-axis conversion by the color correction device according to the first embodiment. In FIG. 3, the color difference signal space is divided by four axes of Cb axis, Cr axis, Cb-axis, and Cr-axis. The axis obtained by rotating the C b axis by a predetermined rotation angle and setting the gain to a primary coefficient of 1! Is the C b 'axis. Further, by rotating the C r axis by a predetermined rotation angle 0 _2, the axis in which the gain as the primary factor 1 _2, and C r 'axis. Furthermore, C b - by rotating the shaft by a predetermined rotation angle theta _3, the axis of the primary coefficient 1 ₃ gain is, C b '- the axis Moreover, C r uniaxial a predetermined rotation angle 0 ₄ The axis rotated by only the first and the gain is set to the primary coefficient 1 is C r 'one axis.

In the 4-axis converter 2, as described above, the 4-axis (Cb-axis, Cb-axis, Cr-axis

, C r -axis) are independently rotated and expanded / contracted to obtain the input color difference signal.

Performs conversion of (C b, C r).

For example, it is assumed that the coordinates of the color difference signal (Cb, Cr) input to the 4-axis conversion unit 2 in the color difference signal space are Q (R, P). In other words, the input color difference signal (Cbr) Is represented by a vector from the origin of the color difference signal space to the coordinates Q (R, P). Since the color difference signal (C b, C r) exists in a region surrounded by one axis of C b where the color difference signal (C b) is negative and the Cr axis where the color difference signal (C r) is positive. The four-axis conversion unit 2 determines that the color difference signal (Cb, Cr) exists in the second quadrant. In the 4-axis conversion unit 2,

, / - depending on the result, performs a conversion using the conversion matrix C ₂ of the formula 3 with刖述. Thus, point R is converted to point U and point P is converted to point S. In other words, after conversion The vector T of the color difference signal (C b ', C r') is defined by two lines: a line on the C b 'axis from the origin to the point U and a line on the C r' axis from the origin to the point S. It is represented by the diagonal of a parallelogram generated as

Here, in the chrominance signal space, the area composed of the four axes (C b '-axis, C b' -axis, C r '-axis, and C r' -axis) converted by the four-axis conversion unit 2 is newly added. Quadrant. In other words, the area consisting of the C b 'and C r' axes is the new first quadrant, and the area consisting of the C r 'and C b' — axes is the new second quadrant, C b '— The area consisting of the axis and the C r '— axis is the new third quadrant, and the area consisting of the C r' — axis and the C b 'axis is the new fourth quadrant.

In FIG. 3, the force ³ ′ rotating all four axes is not limited to this. For example, in FIG. 4, the chrominance signal space is divided by four axes: Cb axis, Cr axis, Cb-axis, and Cr axis. Rotate the C r axis by a predetermined rotation angle 0 _2, the axis of the primary coefficient 1 ₂ gain becomes C r 'axis. Further, the without rotating the C b-axis (rotation angle 6 _»3 = 0), is the axis of the primary coefficient 1 ₃ gain, C b '- the axis. In addition, the C b axis and the C r — axis are not rotated and the gain is set to 0, so that the C b ′ axis and the C r ′ axis coincide with the C b axis and the C r axis, respectively.

For example, it is assumed that the coordinates of the color difference signal (Cb, Cr) input to the 4-axis conversion unit 2 in the color difference signal space are Q (R, P). The color difference signal (C b, C r) exists in the area surrounded by the C b — axis where the color difference signal (C b) is negative and the r axis where the color difference signal (C r) is positive. The four-axis converter 2 determines that the color difference signals (C b, C r) exist in the second quadrant. In 4 axis conversion portion 2, depending on the result, performs a conversion using the conversion matrix C ₂ of the formula 3 described above. Thus, point R is converted to point U and point P is converted to point S. That is, the vector T of the converted color difference signal (C b ′, C r ′) is , Expressed by the diagonal of a parallelogram generated with the line on the C b axis from the origin to the point U and the line on the C r 'axis from the origin to the point S as one side, where 4 axes Assume that the coordinates of the color difference signals (C b, C r) input to the conversion unit 2 in the color difference signal space are B. The color difference signal (C b, C r) exists in an area surrounded by the C b axis where the color difference signal (C b) is positive and the C r axis where the color difference signal (C r) is negative. Therefore, the four-axis conversion unit 2 determines that the color difference signals (Cb, Cr) exist in the fourth quadrant. In the 4-axis conversion unit 2, since the Cb axis and the Cr single axis do not rotate and the gain is 0, the coordinate C of the converted color difference signal (Cb · ', Cr-) is converted. It is the same as the coordinate B of the previous color difference signal (C b, C r). In this way, by not rotating all four axes, it is possible to perform color correction without giving any shadow m to the color of the quadrant constituted by the axes that are not rotated.

Next, a specific example of a method of determining a rotation angle and a gain of a conversion matrix in the color correction device of the first embodiment will be described with reference to the drawings. There are two ways to decide. One is a method in which a user inputs a rotation angle and a gain while referring to an image before color correction and an image after color correction. The other method is to determine the rotation angle and the gain by calculation so that the difference between one or more types of color difference signals and each target color difference signal is minimized.

In order to realize the first method in which the user inputs the rotation angle and the gain while referring to the image before the color correction and the image after the color correction, for example, a tool having a screen as shown in FIG. Is used. This tool is software that runs on a personal computer. FIG. 5 is a diagram showing an example of a screen of a tool for calculating a rotation angle and a gain of a transformation matrix in the color correction apparatus according to the first embodiment. In FIG. 5, the tool screen includes a first display area 11, a second display area 12, a coefficient rotation angle input area 13, and a color difference signal comparison area 1.

4 、 Conversion matrix display area 15 、 Operation execution port 16 Yes.

The first display area 11 1 displays an image before color correction is performed. The second display area 12 displays an image after color correction.

Factor rotation angle input area A 1 3, the respective axes of the color difference signal space, the primary factor of their respective gain (1! ~ 1 ₄₎ and an input for inputting a rotation angle (i-theta 4) It has parts 13a to 13d. Color difference signal comparison area 1

In 4, it is displayed that the color difference signal (Cb, Cr) corresponding to the point A in the image displayed in the first display area 11 exists at the coordinate B.

It is displayed that the converted color difference signal (C b, C r) obtained by color correcting the color difference signal (C b, C r) at point A exists at the coordinate C.

In addition, the conversion matrix display area 15 displays the value of a conversion matrix (expression shown in FIG. 23) determined according to the primary coefficient and the rotation angle input in the coefficient rotation angle input area 13.

In such a case, the user selects a position (point A) in the image displayed on the first display area 11 for performing color correction using an operating device such as a mouse. When point A in the image displayed on the first display area 11 is selected, the color difference signal comparison area 14 displays the point A in the image displayed on the first display area 11. Displays the coordinate B of the corresponding color difference signal (Cb, Cr)

The user confirms that the coordinate B exists in the second quadrant between the C r axis and the C b — axis. For example, the input unit on the C r axis of the coefficient rotation angle input area 13 Enter the primary coefficient 1 ₂ and the rotation angle θ _{2 in} 13 b. Then, the user operates the calculation execution button 16. When the calculation execution button 1 6 is operated, the tools, the second transformation matrix used in quadrant relative to (2 3 (formulas described b)), one entered order coefficient 1 ₂ and the rotation angle 0 ₂ and the product of the color difference signals (C b, C r) is calculated by the equation shown in FIG. Ask. The color difference signal comparison area 14 displays the coordinates C of the converted color difference signals (C b ′, C r ′) after color correction, and the conversion matrix display area 15 inputs the coordinates C to the input section 13 b. the value of the transformation matrix is calculated Me is displayed by the primary factor 1 ₂ and the rotation angle are _two values were. Further, the image after the color correction is displayed in the second display area 12.

The user can adjust the color-corrected image displayed in the second display area 12, the coordinate B of the color-difference signal (C b, C r) displayed in the color-difference signal comparison area 14, the converted color difference Look at the coordinates C of the signals (C b ', C r') and the values of the conversion matrix displayed in the conversion matrix display area 15, and look at the coefficient ₁₂ and the rotation angle input to the input section 13 b. 0 ₂ values can be confirmed whether or not appropriate. Then, if inappropriate, re-enter the rotation angle and gain to make adjustments.

In the tool shown in FIG. 5, the rotation angle and the gain of the transformation matrix are obtained by inputting a numerical value into the coefficient rotation angle input area 13, but the present invention is not limited to this. For example, the color difference signal space is displayed in the coefficient rotation angle input area 13 and axes in the color difference signal space (Cb-axis, Cr-axis, Cb-axis, Cr-axis) are displayed using an operating device such as a mouse. ) May be directly rotated or expanded or contracted.

In addition, in order to realize a second method of determining the rotation angle and the gain so that the difference between each target color difference signal with respect to two or more types of color difference signals is minimized, The following calculation is performed. First, since there are two or more types of color difference signals (C b, C r) before color correction, the i-th color difference signal (C b, C r) is defined as (C bi, C ri) (i Is an integer greater than or equal to 1). Also

, 変換₂ , Θ 3, 0 ₄ , 1!, 1 ₂ , 1 3

, 1 _4), and to each of the color difference signals (C bi, the color difference signal to be the target of C ri) (C b ', C r' a) and (C bi C ri '). And Figure 2 Set the equation shown in ₄ and calculate 0 i Θ ₂ , θ 3, θ, 1 or 1 ₂ 1 3-1 4 that minimizes g. In other words, the sum of the difference between the color difference signal before conversion (C bi C ri) and the color difference signal after conversion (C bi C ri) is the smallest. 0 ₂ , Θ 3, Θ 4 1] L 1 Calculate ₂ 1 3 '1 ₄

g minimum to theta i theta a 1 θ 3 Θ 4 1:. L> 1 2, 1 3 - 1 4 as how to calculate the the gradient method or the simplex method, Aniri ring method, GA

(Genetic algorithm) can be used. As the color difference signals (C b, Cr) before color correction, for example, the color west where the spectral reflection characteristics of the color filters, etc., are separated is known. It is possible to use signals obtained by imaging

As described above in detail, according to the first embodiment, the four axes (Cb axis, Cb-axis Cr axis, and Cr single axis) in the color difference signal space are independently rotated and expanded / contracted. Shift the coordinate point of the input color difference signal (CbCr).

Since color correction is performed, when one or two of the four axes in the color difference signal space are rotated and expanded / contracted to correct one color, the effect on other colors is small. The other axes can be rotated and expanded / contracted so that the desired color is color-corrected, thereby reducing the influence on other colors. Also, by rotating and expanding one or two of the four axes in the color difference signal space to correct one color, the other axes do not change when the other axes It is also possible to correct only a specific color without giving a color.

In addition, by independently rotating the four axes, the colors on both sides of the axis are converted following the rotation of the axis, so that the continuity of the colors at the axis boundaries can be easily maintained without any special consideration. Therefore, color correction with a high degree of freedom can be easily performed, and color reproducibility as a whole can be improved. In addition, since color correction is performed using the color difference signal space, Makes it easier to imagine the color after conversion, and makes color correction intuitive.

(Second embodiment)

Hereinafter, a second embodiment of the present invention will be described with reference to the drawings. FIG. 6 is a block diagram illustrating a configuration example of a color correction device according to the second embodiment.

. In FIG. 6, reference numeral 21 denotes a luminance / chrominance signal separation unit, which receives an input image signal.

The luminance signal (Y) and the color difference signals (Cb, Cr) are separated and output.

Reference numeral 22 denotes a clogan conversion unit which receives the color difference signals (Cb, Cr) separated by the luminance / color separation unit 21 and outputs the color difference signals (Cb, Cr).

) Is converted according to a predetermined gamma force curve to generate gamma-converted color difference signals (C b ′ ′, C r ′ ′).

The gamma force is represented, for example, by a graph shown in FIG. The color difference signal (C b) is converted into a color difference signal (C b

Is converted to ''). The color difference signal (Cr) is converted into a color difference signal (Cr '") after gamma conversion by a gamma curve G2. In the color correction device according to the second embodiment, the gain of the color difference signals (Cb, Cr) is obtained from the gamma curves Gl, G2. The gamma curves G l and G 2 can be freely set using tools and the like.

Reference numeral 23 denotes a four-axis conversion unit which receives the gamma-converted color difference signal (Cb'Cr) converted by the chroma gamma conversion unit 22 and whose coordinates correspond to which quadrant (color space) in the color-difference signal space. It determines whether it is included in any of the first to fourth quadrants), and calculates the product of the conversion matrix determined according to this and the color difference signals (C b '', C r '') after gamma conversion Thus, the converted color difference signals (C b ′, C r ′) obtained by color correction of the color difference signals (C b, C r) are obtained. If the coordinates of the color difference signals (C b '', Cr '') after gamma conversion are in the first quadrant, color correction is performed using the conversion matrix C i shown in Fig. 25 (a). . Similarly, the coordinates of the color difference signals (C b '', C r '') after gamma conversion are When present in limited performs by Ri color correction conversion matrix C ₂ shown in FIG. 2 5 (b). If the coordinates of the color difference signals (C b '', C r '') after gamma conversion are in the third quadrant, color correction is performed using the conversion matrix C ₃ shown in Fig. 25 (c). If the coordinates of the color difference signal (Cb '' Cr '') after gamma conversion exist in the _fourth quadrant, color correction is performed using the conversion matrix C4 shown in Fig. 25 (d).

Ό ·

Reference numeral 24 denotes a luminance / chrominance signal component, which is separated by the luminance / chrominance signal separation unit 21 from the converted color difference signal (CbCr ') subjected to color correction by the 4-axis conversion unit 23. The image signal is generated by combining the obtained luminance signal (Y) and, and output to the outside.

FIG. 8 is a diagram illustrating an example of four-axis conversion by the color correction device according to the second embodiment. In the example shown in FIG. 8, the color difference signal (CbCr) existing at the coordinate Q in the second quadrant and the color difference signal (CbCr) existing at the coordinate B in the fourth quadrant are converted. The chrominance signals (C b,

C r) is gamma-converted by the Ku-gamma converter 22. The gamma force G 1 G 2 used at this time is as shown in Fig. 9.

. In FIG. 9, when the negative value of the Cb component of the color difference signal increases, the negative value of the Cb ′ ′ component of the gamma-converted color difference signal increases as the negative value of the Cb component of the color difference signal increases. It is set to be larger than <. Also the gamma curve

G 2 is set so that the Cr component of the color difference signal is equal to the Cr ′ ′ component of the color difference signal after gamma conversion.

The color difference signal (C b

C r ′) is output to the 4-axis conversion unit 23. The four-axis conversion unit 23 performs color correction using a conversion matrix corresponding to the quadrant where the color difference signal after the gamma conversion exists. That is, the color difference signals (C b C r) of the coordinates Q are gamma-converted and the color difference signals (C b ′, C r ′) are converted to the conversion matrix C ₂ shown in FIG. 25 (b). The color difference signal (C b, Cr) obtained by gamma-converting the color difference signal (C b, Cr) at coordinate B is converted as shown in Fig. 25 (d). perform by Ri color correction matrix C _4.

Therefore, the color difference signal (Cb, Cb) obtained by converting the color difference signal (Cb, Cr) of the coordinate Q

, C r ′) becomes T, and the coordinates of the color ¾i Is->, C b 1, cr ′) obtained by converting the color difference signal (C b, C r) at coordinate B become C. Thus, for example, while increasing the saturation of the yellow color at the coordinate Q and correcting the color to the coordinate T at which the redness has been reduced, the green color of the coordinate B can be increased while maintaining the saturation. The color can be corrected to the coordinates C.

As described in detail above, according to the second embodiment, the color difference signals (C b, Cr) are converted by the gun force G 1, G 2, and the color difference signals (gamma converted) C b '' .. C r '') is converted to 4-axis, so the color difference signal (C b

This set the gain fin to the C r) instead of the primary coefficient 1 E to 1 _4: and leave in. At this time, depending on the setting of the gamma force Gl, G2, the original color difference signal

The size of the gain can be varied according to the size of (C b, C r). This makes it possible to apply gain to the color difference signals (Cb, Cr) in the same area unevenly. Therefore, the color difference signal in the same area

(C b, C r) can be easily subjected to color correction with a degree of freedom other than 搽 and color reproducibility can be improved.

By using the equation shown in FIG. 23 instead of the equation shown in FIG. 25 in the 4-axis conversion unit 23, it is possible to further apply a gain by the following coefficients 1 to 1.

(Third embodiment)

Hereinafter, a third embodiment of the present invention will be described with reference to the drawings. FIG. 10 is a block diagram illustrating a configuration example of a color correction device according to the third embodiment. In FIG. 10, reference numeral 31 denotes a luminance / chrominance signal separation unit, which receives an input image. The luminance signal (Y) and the color difference signals (Cb, Cr) are separated from the signal and output.

Reference numeral 3 2 denotes a four-axis conversion unit which receives the color difference signals (C b, C r) separated by the luminance / color difference signal separation unit 31 and converts the color difference signals (C b, C r) into two. A color difference signal (Cb, Cr, Cb2, Cr2) containing the following terms is generated. Here, C b 2 = C b ′ (| C b | + | C r |), and C r 2 = C r · (IC b I + IC r I). Further, the 4-axis conversion unit 32 determines which quadrant (any one of the first to fourth quadrants) in the color difference signal space includes the coordinates of the input color difference signal (C b, Cr). Then, the product of the transformation matrix C _n (n is an integer from 1 to 4) determined according to this and the color difference signals (C b, C r, C b 2, C r 2) including the quadratic terms is plotted. The color difference signals (Cb ', Cr') after color correction of the color difference signals (Cb, Cr) are obtained by calculating using the formula shown in 26 (a). Where the transformation matrix C _n is the rotation angle of the four axes

Θ丄~ θ _{4, 4} primary factor that indicates the gain of the shaft 1 to 1 ₄ and secondary coefficient! !!!! ~! !!!!

Including ₄ .

If the coordinates of the color difference signal (C b, C r) are in the first quadrant, color correction is performed using the conversion matrix C i shown in FIG. 26 (b). The coordinates of the color difference signals (C b 'CV) is when present in the second quadrant performs color correction by conversion matrix C ₂ shown in FIG. 2 6 (c). If the coordinates of the color difference signal (C b, C r) exist in the third quadrant, color correction is performed using the conversion matrix C ₃ shown in FIG. 26 (d), and the color difference signal (C b, C r) If the coordinates are present in the fourth quadrant of) performing color correction by the transformation matrix C ₄ shown in FIG. 2 6 (e).

Reference numeral 3 denotes a luminance / chrominance signal synthesizing unit, which is separated by the luminance / chrominance signal separating unit 3 1 from the converted color difference signals (C b ′, Cr ′), which have been color-corrected by the 4-axis conversion unit 3 2. The image signal is generated by combining the luminance signal (Y) and output to the outside. Next, a specific example of four-axis conversion by the color correction device according to the third embodiment will be described with reference to the drawings. FIG. 11 is a diagram illustrating an example of four-axis conversion by the color correction device according to the third embodiment. In FIG. 11, the color difference signal space is divided by four axes: Cb axis, Cr axis, Cb-axis, and Cr-axis. The axis in which the C b axis is rotated by a predetermined rotation angle Θ i, and the gain is represented by the primary coefficient 1 and the secondary coefficient is the C b ′ axis. Further, by rotating the C r axis by a predetermined rotation angle 0 _2, the axis of the the gain as the primary factor 1 ₂ and the secondary coefficient m ₂ is a C r 'axis. Furthermore, C b - by rotating the shaft by a predetermined rotational angle theta _3, is the axis of the primary factors 1 ₃ and the secondary coefficient m ₃ a gain, C b '- the axis. Also, C r - by rotating the shaft by a predetermined rotational angle theta _4, the axis of the the gain as the primary coefficient of 1 ₄ and the secondary coefficient m ₄ is, C r '- the axis.

As described above, the four-axis conversion unit 32 independently rotates and expands and contracts the four axes (Cb axis, Cb-axis, Cr axis, and Cr single axis) as described above, so that the input color difference signal (C b, C r).

For example, as shown in FIG. 12, it is assumed that the coordinates of the color difference signals (C b, C r) input to the four-axis conversion unit 32 in the color difference signal space are Q (R, P). That is, the input color difference signal (C b, C r) is represented by a vector from the origin of the color difference signal space to the coordinate Q (R, P). In other words, the vector representing the input color difference signal (C b, C r) is a rectangle formed by the line on the C b axis from the origin to the point R and the line on the Cr axis from the origin to the point P. It is represented by the diagonal of.

This input color difference signal (C b, C r) exists in the area surrounded by the C b -axis where the color difference signal (C b) is negative and the Cr axis where the color difference signal (C r) is positive. Therefore, the four-axis conversion unit 32 determines that the color difference signals (C b, C r) exist in the second quadrant. In accordance with this result, the 4-axis conversion unit 32 uses the conversion matrix C ₂ expressed by the above-described equation shown in FIG. 6 Conversion is performed using the equation shown in (a). Because it contains secondary coefficient m ₂ m ₃ the transformation matrix C _2, as compared with the case of only the primary factor 1 ₂ 1 _3, gain is increased when the value of m _2, m ₃ is positive, When the value of m ₂ m ₃ is negative, the gain becomes small. As a result, the converted color difference signal (Cb'Cr ') is divided into a line from the origin on the Cb'-axis to point U and a line from the origin on the Cr' axis to point S on the Cr 'axis. It is represented by the vector up to the coordinate T 'on the extension of the diagonal line (the vector from the origin to the coordinate T) of the parallelogram generated as two sides. Note that vector from the origin to the coordinates T is the first Ri by the color correction device of the embodiment color-corrected base-vector of the color difference signal using only the primary coefficient 1 ₂ 1 _3.

As described in detail above, according to the third embodiment, the color difference signal (C b, C r) is converted into a form (C b, C r, C b 2, C r 2) including a quadratic term To the EiX Ah of the 5 積 -coefficient, since color correction is performed by calculating the product of the conversion matrix with the primary coefficient (lil ₄ ) and the secondary coefficient (ni im A). The degree of freedom increases. In other words, it is possible to set a coefficient so that the gain increases as the distance from the origin increases, and to set a coefficient such that the gain decreases as the distance increases from the origin. For example, when performing yellow color correction, the saturation of yellow is increased.

However, it is possible to perform conversion so as not to increase the saturation of yellow with a low saturation.0 Therefore, it is possible to easily perform color correction with a higher degree of freedom. 0

(Fourth embodiment)

Hereinafter, a fourth embodiment according to the present invention will be described with reference to the drawings. Figure 1

3 is a block diagram showing a configuration example of a color correction device according to the fourth embodiment.o In FIG. 13, reference numeral 41 denotes a luminance / color signal separation unit, which is a part of the luminance / color signal separation unit. The luminance signal (Y) and the color difference signals (Cb, Cr). The

Reference numeral 42 denotes an 8-axis conversion unit (corresponding to the axis conversion unit in the claims), which converts the color difference signals (C b, Cr) separated by the luminance / color difference signal separation unit 41, A color difference signal (C b, C r, C b 2, C r 2) including a quadratic term is generated. Also, the 8-axis conversion unit 42 determines in which area the coordinates of the input color-difference signal (Cb, Cr) are present in the color-difference signal space in which the area is equally divided by the eight axes. Then, a transformation matrix c _n (n is an integer of 1 to 8) determined according to this and a color difference signal (C b, C r, C

2, C r 2) to obtain the converted color difference signals (C b ′, C r ′) after color correction of the color difference signals (C b, cr),

As a method of determining in which page area of the eight areas the coordinates of the input color difference signal (C b, C r) are located, for example, in the color difference signal space of the color difference signal (C b, C r) There is a method of calculating the pole from the coordinates of the coordinates and comparing the calculated angle information of the polar coordinates with the angle information of the eight axes.

Reference numeral 43 denotes a luminance / chrominance signal synthesizing unit D. The luminance / chrominance signal separating unit 41 converts the color difference signals (Cb ', Cr'), which have undergone color correction by the 8-axis conversion unit 42, and the luminance / chrominance signal separation unit 41.

Generates an image signal by combining the luminance signal separated by

Next, a specific example of 8-axis conversion by the color correction device of the fourth embodiment will be described with reference to the drawings. FIG. 14 is a diagram illustrating an example of 8-axis conversion by the color correction device according to the fourth embodiment. In Figure 14, the color difference signal space is

, Cr axis, C b -axis, C r One axis is divided equally into four axes of K axis, L axis, M axis and N axis which are in the middle of these four axes. Is

. For example, an axis obtained by rotating the axis by a predetermined rotation angle Θi and setting the gain to a −first order 1 i and a second order coefficient mi is the C b ′ axis. Also, place the K axis It is rotated by a constant rotation angle theta _2, the axis of the primary factors 1 ₂ and the secondary coefficient m ₂ the gain is a K 'axis.

For example, as shown in Fig. 14, the color difference signal input to the 8-axis conversion unit 42

The vector V of (C b, C r) is a parallelogram formed by a line from the origin on the C b axis of the color-difference signal space to the point W on the W axis and a line from the origin on the K axis to the point X. Is represented by a diagonal line. In the 8-axis conversion unit 42, the color difference signals (C b, C r

) Is located in the area between the Cb and K axes, so color correction is performed using the formula shown in Fig. 27. As a result, the converted color difference signal (C b 'C r

One

The vector V in (1) is a pair of parallelograms generated with two lines, the line from the origin on the C b 'axis to the W — point and the line from the origin on the K' axis to the X 'point. Here, the W 'point on the Cb' axis is represented by an angle, and the W point is converted by rotating the Cb axis by Θ and expanding and contracting by the primary coefficient 1i and the secondary coefficient mi. point 'X on the axis' are those the K rotates the K axis by theta _2, it is obtained by converting the X point by stretching the primary factor 1 ₂ and the secondary coefficient m _2.

As described in detail above, according to the fourth embodiment, since the color signal space is divided into eight axes and the conversion is performed, finer color correction can be performed. Further, for example, a color difference signal (CbCr) existing between the Cb axis and the κ axis.

) Color correction □, C b axis and K axis rotate and expand and contract, between C axis and L axis, between L axis and C b — axis, between C b — axis and M axis , M-axis and Cr-axis, Cr Color difference signal (Cb

Since C r) is not affected, it is possible to further reduce the effect on other colors when a specific color is corrected, as compared with the case where conversion is performed by dividing by four axes. it can. Therefore, it is possible to perform color correction with a higher degree of freedom and to improve color reproducibility. In the fourth embodiment, the color difference signal space is represented by the Cb axis and the Cr axis. C b-axis C r — axis 4 axis, and K axis, L axis,

It is composed of a total of 8 axes including the 4 axes of M axis and N axis, but is not limited to AE.For example, 12 axes, 16 axes, etc. The region of the color difference signal space may be divided.

Moreover, as shown in FIG. 15, the positions of the axes in the color difference signal space may be freely arranged without being determined in advance. According to Fig. 15, between the Cb axis and the Cr axis, the E axis rotated by the rotation angle from the Cb axis and the F axis rotated by the rotation angle 1 from the C axis are provided, and the E axis is only だけ. rotate, respectively co ί rotate the F axis by theta ₂ - is the next coefficient 1, 1 by 2 and the secondary coefficient mm ₂ so as to stretch the E-axis and F axis. In this case, the input color signal (C b, C r) is converted to generate β ¾Ξ 15 (C b, C r C b 2, C r 2) including the quadratic term, and The converted color difference signals (C b ′, C r ′) are obtained by the equation shown in FIG.

For example, when converting a color difference signal (C b, C r) existing between the C b axis and the C r axis, the color signal (C b C r) must be located on the E axis or F axis. Then, the E-axis and the F-axis may be rotated and expanded / contracted. This makes it possible to directly correct the color difference signals (Cb, Cr) to be corrected by rotating and expanding and contracting the E-axis and F-axis. Therefore, color correction with a higher degree of freedom can be easily understood <easy to perform and color reproducibility can be improved. In addition, the position of the axis in the color difference signal space can be freely determined without being determined in advance. And the number of areas divided by the axis is N (

(N is an integer of 1 or more). When arbitrary coordinates on each axis are set, the axes may be set so that the outer product of adjacent coordinates is all larger than 0. For example, as shown in FIG. 17, assume that the number of divided areas is 5, and the axes that divide the areas are 0 axis, 1 axis, 2 axes, 3 axes, and 4 axes. And any coordinates on the 0 axis (AO x, A 0 y), any coordinates on the 1 axis (A 1 X, A 1 y), and any coordinates on the 2 axis (A 2 X, A 2 y ), Set arbitrary coordinates (A 3 X, A 3 y) on three axes, and set arbitrary coordinates (A 4 x, A 4 y) on four axes. Here, the cross product of adjacent coordinates on the 0 axis (AO x, A 0 y) and coordinates on the 1 axis (A 1 X, A 1 y) is (AO x, AO y) x (A 1, A 1 y) = A 0 x-A ly-AO y · Α 1 x. Similarly, the outer product of the coordinates (A 1 X, A 1 y) on one axis and the coordinates (A 2 X, A 2 y) on the two axes that are adjacent to each other is (A lx, A ly) X (A 2x'A2y) = A1X • A2y-A1y, A2x. The cross product of adjacent coordinates on two axes (A 2 X, A 2 y) and coordinates on three axes (A 3 x, A 3 y) is (A 2 XA 2 y) X (A 3 x , A 3 y) = A 2 x-A 3 y-A 2 y · A 3 x. The cross product of adjacent coordinates on three axes (A 3 x, A 3 y) and coordinates on four axes (A 4 x, A 4 y) is (A 3 x, A 3 y) X ( A4x, A4y) = A3xA4y—A3y'A4x. The outer shell of the coordinates (A 4 x, A 4 y) on the adjacent 4 axes and the coordinates (AO x, A 0 y) on the 0 axis is (A 4 x, A 4 y) X (AO x, A 0 y) = A 4 x · A 0 y — A 4 y • A 0 x. Then BX Ah each axis so that all cross products are greater than zero. When each axis is set as described above, the axis conversion unit can determine in which area the coordinates of the color difference signal (Cb, Cr) exist. Specifically, the cross product of the coordinates of the color difference signal (Cb, Cr) and the coordinates on the 0 axis (AOx, A0y) is 0 or more, and the coordinates of the color difference signal (CbCr) If the cross product of the coordinates on the first axis and (A 1 X, A 1 y) is smaller than 0, the color difference signal (C b, C r) is the area between the 0 axis and the 1 axis. Is determined to exist. The coordinates of the color difference signals (C b, Cr) and the coordinates on one axis (A 1 X, A 1 y) is 0 or more, and the coordinates of the color difference signals (C b, C r)

If the cross product with the coordinates (A 2 X, A 2 y) on the two axes is smaller than 0, it is determined that the color difference signal (C b C r) exists in the area between the 1 and 2 axes I do. In addition, the coordinates of the color difference signals (Cb, Cr) and the coordinates on two axes (A2X,

A 2 y) is 0 or more, and the coordinates of the color difference signals (C b, C r)

If the cross product with the coordinates (A3X, A3y) on the three axes is smaller than 0, it is determined that the color difference signal (CbCr) exists in the area between the two axes and the three axes I do. Also, the coordinates of the color difference signals (Cb, Cr) and the coordinates on three axes (A3X,

A 3 y) is 0 or more, and the coordinates of the color difference signal (C b, C r)

If the cross product with the coordinates on the four axes (A4X, A4y) is smaller than 0, the color difference signal (Cb, Cr) is considered to exist in the area between the three axes and the four axes. Judgment 1 9 First, the coordinates of the color difference signals (Cb, Cr) and the coordinates on four axes (A4X,

A 4 y) is greater than or equal to 0 and the coordinates of the color difference signals (C b, C r)

If the cross product with the coordinates on the 0 axis (A0X, A0y) is smaller than 0, it is determined that the color difference signal (CbCr) exists in the area between the 4th axis and the 0th axis I do.

Also, by using a screen with a screen as shown in Fig. 16, it is possible to add an axis passing through the coordinates of the color signal (Cb, Cr) to be corrected and rotate and expand / contract that axis. You may do it. FIG. 16 is a diagram showing a screen example of a tool for calculating a rotation angle and a gain of a transformation matrix when adding an axis in the color correction apparatus according to the fourth embodiment. In FIG. Is

, First display area 11, second display area 12, coefficient rotation angle input area 17, color signal coordinate angle display area 18, axis addition button 19, color difference signal comparison area 14, a conversion matrix display error 15, and an operation execution button 16.

The first display area 11 displays an image before color correction is performed. The second display layer 12 displays the image after color correction.

The coefficient rotation angle input area 17 has an axis number indicating the number of each axis in the color difference signal space and an area indicating the angle of each axis. Is the angle based on In the coefficient rotation angle input area 17, the primary coefficient (111) and the secondary coefficient (mlm4) indicating the gain for each axis of the color difference signal space, and the rotation of each axis are shown. Has an area for inputting a rotation angle (0104) indicating the angle to be set. Also, the axis of axis number 14 is an axis prepared in the initial state, and the axis of axis number 5 is , Added by the axis addition button 19. When an axis is added by operating the axis addition button 19, an area indicating the axis number and angle of the axis, and an area for inputting a primary coefficient, a secondary coefficient, and a rotation angle are added.

The color difference signal 角度 τη. Angle display area 18 is the first display area 11 1t; the coordinates and origin of the color difference signal (CbCr) at point Α in the displayed image are The angle of the connecting axis is displayed.The axis addition button 19 is a button for adding the axis of the angle displayed in the color difference signal coordinate angle display area 18 as the axis of the color difference signal space.

In the color difference signal comparison area 14, it is displayed that the color difference signal (C b, C r) corresponding to A in the image displayed in the first display area 11 exists at the coordinate B ₀ † The color difference signal at point A (CbCr) is color corrected.

-It is displayed that the converted color difference signal (Cb, Cr ') exists at the coordinate C.

The transformation matrix display area 15 displays the value of the transformation matrix determined according to the rotation angle, the primary coefficient, and the secondary coefficient input in the coefficient rotation angle input area 17.

In such a tool, the user uses an operating device such as a mouse to Select the position (point A) in the image displayed in the first display area 11 where color correction is to be performed. When point A in the image displayed in the first display area 11 is selected, the color difference signal comparison area 14 displays the point A in the image displayed in the first display area 11. The coordinates B of the color difference signal (Cb, Cr) corresponding to are displayed. In the color difference signal coordinate angle display area 18, the angle of the axis connecting the coordinates of the color difference signal corresponding to the point A in the image displayed in the first display area 11 and the origin is displayed. .

The user decides whether or not to add the axis of the angle displayed in the color difference signal coordinate angle display area 18 as an axis of the color difference signal space. When adding, the axis addition button 19 To operate. When the axis addition button 19 is operated, the coefficient rotation angle input area 17 shows the C b axis (axis number 1, angle 0 °, rotation angle 0 °) that originally exists. None), Cr axis (axis number 2, angle 90 °, rotation angle 0 °, no primary and quadratic coefficients), Cb axis (axis number 3, angle 180 °, rotation angle 0 °, In addition to the four axes of primary axis (without primary and secondary coefficients), Cr and one axis (axis number 4, angle 270 °, rotation angle 0 °, no primary and secondary coefficients), axis number 5, angle A 120 ° axis is added. Then, the user inputs values of the rotation angle 0 ₅ , the primary coefficient ₁₅ , and the secondary coefficient m ₅ of the axis number 5 of the coefficient rotation angle input area ₁₇ , and operates the calculation execution button 16.

When the operation execution button 16 is operated, the tool converts the values of the input rotation angle 0 ₅ , first-order coefficient ₁₅ , and second-order coefficient m ₅ to the transformation matrix used in the second quadrant by C The product is calculated by taking into account the difference in angle from the r axis and the color difference signal. The color difference signal comparison area 14 displays the coordinates C of the color-corrected color difference signals (C b ′, C r ′) after conversion. The conversion matrix display area 15 displays the input rotation angle. The value of the transformation matrix obtained from the values of 0 ₅ , primary coefficient ₁₅ , and secondary coefficient m ₅ is displayed. Also, the second display area 1 2 shows the image after color correction.

The user can adjust the coordinates B of the color-corrected image displayed in the second display area 12, the coordinates B of the color-difference signals (C b, C r) displayed in the color-difference signal comparison area 14, Look at the coordinates C of the color difference signals (C b ', C r') and the value of the conversion matrix displayed in the conversion matrix display area 15, and look at the rotation angle 0 input to the coefficient rotation angle input area 17. _5. It is possible to confirm whether the values of the primary coefficient ₁₅ and the secondary coefficient ms are appropriate. If not, re-enter and make adjustments. In the fourth embodiment, a quadratic coefficient is used, but only a first-order coefficient may be used.

Note that, in the color correction devices according to the third to fourth embodiments,

, A gun conversion unit may be provided to perform conversion by gun force.

(Fifth embodiment)

Hereinafter, a fifth embodiment according to the present invention will be described with reference to the drawings. Figure 1

FIG. 8 is a block diagram illustrating a configuration example of a color correction device according to a fifth embodiment. In FIG. 18, reference numeral 51 denotes an RGB matrix U / X conversion unit which receives an RGB signal (R, G, B) and calculates a product of the RGB signal and the 3 × 3 conversion matrix shown in FIG.

One

To obtain the converted RGB signals (R, G, B ').

As a conversion matrix for converting the RGB signal (R, G, B), a value predetermined by a parameter such as software for performing color correction is used.

Reference numeral 52 denotes a color difference signal extraction unit that extracts a color difference signal (Cb, Cr) corresponding to the first-order term from the input RGB signals (R ', G', B ') and outputs the signal. The color difference signal extraction section 52 extracts a color difference signal (Cb2, Cr2) corresponding to a quadratic term from the RGB signal (R ', G', B ') and outputs the signal.

Reference numeral 53 denotes a first conversion matrix generation unit which receives a color difference signal (C b, C r) corresponding to a first-order term, and generates a first conversion matrix A based on this. here Then, the first transformation matrix A is obtained by the product of the following three matrices. The three matrices are used to convert the input converted RGB signals (R ', G', B ') to obtain a luminance signal (Y) and a color difference signal (Cb, Cr). queue

(For example, the matrix shown in FIG. 30 (a)) and a second color difference signal (C b ′, C r) obtained by converting the input color difference signal (C b, C r) to obtain a color difference signal (C b ′, C r) Matrix (for example, the matrix shown in Figure 30 (b)) and the color-corrected color difference signal

This is the third matrix (for example, the matrix shown in FIG. 30 (c)) for converting (C b ′, C r ′) into RGB signals (R ′, G ′, Β ′). In addition,

1 matrix and the third matrix, c ⁰ lame such software Bok Wea performing color correction

-It depends on the evening. The third matrix is the inverse of the first matrix. In the second matrix, the area is divided by the axis.

In the color difference signal space, the coordinates of the input color difference signals (C b, C r) are determined in which area, and a matrix determined accordingly (for example, FIG. 23 or FIG.

The matrix shown in 25). If the axes are freely arranged as shown in Fig. 15 without predetermining the positions of the axes in the color difference signal space, the second matrix is as shown in Fig. 31 (a). It becomes a matrix.

By calculating the product of the first to third matrices (for example, the equation shown in FIG. 32), the first transformation matrix A is obtained. Here, the first transformation matrix A determines in which area the coordinates of the input color difference signal (C b, C r) exist in the color difference signal space in which the area is divided by the axis. It is a matrix that is determined and determined accordingly. Therefore, the first transformation matrix A exists for the number of regions.

Reference numeral 54 denotes a first image conversion unit, which is a matrix conversion RGB signal (R ′, G ′, B ′) input from the RGB matrix conversion unit 51 and a first conversion matrix A To obtain RGB signals (R ''',G''', B ''') converted from the first-order terms. Reference numeral 55 denotes a second conversion matrix generation unit which inputs a color difference signal (C b 2, C r 2) corresponding to a quadratic term, and generates a second conversion matrix B based on this. Here, the second transformation matrix B is obtained by the product of the following two matrices. The two matrices are used to transform the input color difference signals (C b 2, Cr 2) to obtain color corrected color difference signals (C b ′, Cr ′). In order to convert the color difference signals (C b '', C r) into the RGB signals (R '', G '''', B '''') Is the fifth matrix (for example, the matrix shown in Fig. 33 (b)). The fifth matrix is determined by parameters such as software for performing color correction. Further, the fourth matrix determines in which area the coordinates of the input color difference signal (C b 2, Cr 2) exist in the color difference signal space in which the area is divided by the axis. (For example, the matrix shown in Fig. 26). If the axes are freely arranged as shown in Fig. 15 without predetermining the positions of the axes in the color difference signal space, the fourth matrix is as shown in Fig. 30 (b). It becomes a matrix.

By calculating the product of the fourth to fifth matrices (for example, the equation described in FIG. 34), a second transformation matrix B is obtained. Here, the second transformation matrix B is based on the chrominance signal space in which the region is divided by the axis, and in which region the coordinates of the input chrominance signal (Cb2, Cr2) exist. It is a matrix determined according to this. Therefore, there are as many second transformation matrices B as the number of regions.

Reference numeral 56 denotes a second image conversion unit, and the second conversion matrix B input from the second conversion matrix generation unit 55 and the color difference signals (C b 2, C r 2) input from the color difference signal extraction unit 52 ) To obtain the RGB signal (R ''', G, B''') for transforming the quadratic term. The second image converter 56 converts the first-order terms input from the first image converter 54. Color correction by adding the RGB signals (R '''', G '''' B '''-) for converting the quadratic terms to the RGB signals (R''' GB ') Find RGB signal (RG '''',

As described in detail above, according to the fifth embodiment, it is possible to directly perform color correction on an RGB image signal, so that it is possible to improve the calculation accuracy and to reduce the emulation error. Can be reduced

The color reproducibility can be improved.

According to the fifth embodiment, a conversion matrix for converting RGB signals (R, GB) (for example, a conversion matrix shown in FIG. 29), and a unit matrix as a first conversion matrix A By using a zero matrix as the transformation matrix B of (2), RGB image signals can be output without color correction.

However, since it is possible to output the RGB image signal as it is without converting it to a color difference signal, there is no calculation difference that occurs when the image signal is converted to an RGB image signal '-J after being converted to a color difference signal.

Further, according to the fifth embodiment, the color difference signal (C b, C r) is converted into a form (C b, C r C b 2, C r 2) including a quadratic term, and the primary coefficient (1 Since the color correction is performed by calculating the product of (a) a) and the transformation matrix in which the quadratic coefficients (m, m2) are set, the degree of whitening for the setting of the coefficients increases. That is, the coefficient can be set so that the gain increases as the distance from the origin increases, or the coefficient can be set so that the gain decreases as the distance from the origin increases. Therefore, it is possible to easily perform color correction with a higher degree of freedom and to improve color reproducibility.

Note that, in the fifth embodiment, the first conversion matrix generation unit 53 and the first image are not provided without the second conversion matrix generation unit 55 and the second image conversion unit 56. Only the primary term color correction by the image converter 54 may be performed. Further, the second correction matrix B generated by the second conversion matrix generation unit 55 may be a zero matrix to nullify the color correction of the quadratic term.

In addition, the above-described first to fifth embodiments are merely examples of specific embodiments for carrying out the present invention, and the technical scope of the present invention is interpreted in a limited manner. It must not be. That is, the present invention can be embodied in various forms without departing from its spirit or its main features. Industrial applicability

INDUSTRIAL APPLICABILITY The present invention is useful for a color correction device for performing color correction of an image signal in an imaging device such as a digital camera, a TV camera, and a video camera, and a display device that captures and displays an image signal. .

Claims

The scope of the claims

1 port A luminance and chrominance signal separation unit that receives an image signal and separates it into a luminance signal and a chrominance signal.

The color signal separated by the luminance color difference signal separation unit is input, and in which region of the plurality of regions divided by a plurality of axes starting from the origin of the color difference signal space does the B color difference signal exist? And an axis conversion unit that converts the color difference signal using a conversion matrix corresponding to the determined area;

A luminance / chrominance signal synthesizing unit for synthesizing the color difference signal converted by the axis conversion unit and the spleen degree signal separated by the luminance / chrominance signal separation unit;

,

The conversion matrix is a matrix having, as elements, a coefficient for independently expanding and contracting the plurality of axes and a rotation angle for independently rotating the R′J B ≦ t number of axes. Color correction

2 mouths

A luminance / chrominance signal / separation unit that receives an image signal and separates the luminance / chrominance / chrominance / chrominance signal

A color signal separated by the luminance color signal separation unit, and a gamma conversion unit for converting the HU Sl5 color difference signal to gamma based on a predetermined gamma curve; and a gamma conversion unit for chroma gamma conversion The obtained chrominance signal is input, and it is determined which area of the plurality of areas divided by the plurality of axes starting from the origin of the chrominance signal space contains the gamma-converted chrominance signal. An axis conversion unit for converting the gamma-converted color difference signal using a conversion matrix corresponding to the region

,

a luminance / chrominance signal synthesizing unit for synthesizing the chrominance signal converted by the m-axis conversion unit and the luminance signal separated by the luminance / chrominance signal separation unit, wherein the conversion matrix includes the plurality of axes. Rotation angle to rotate independently element A color correction apparatus characterized in that

(3) The color correction device according to (1), wherein the plurality of axes are the four axes.

4 multiple axes are 4 at

3. The color correction device according to claim 2, wherein the color correction device is an axis.

5. The color correction device according to claim 1, wherein the plurality of axes are eight axes.

6.The color correction apparatus according to claim 2, wherein the plurality of axes are eight axes.

(7) a chroma-gamma conversion unit that performs gun-to-conversion on the color difference signal based on a predetermined gamma force;

,

The color correction device according to claim 1, wherein the axis conversion unit converts the color difference signal that has been permanently replaced by the FJIJ conversion unit.

8. The color difference signal has a first-order term and a second-order term, and the coefficient is composed of a first-order coefficient for the m-th order term and a second-order coefficient for the second-order term. The color correction device described in Paragraph 1

9. A color difference signal extraction unit for inputting an image signal and extracting a color signal;

The color difference signal extracted by the color difference signal extraction unit is input, and it is determined in which of a plurality of regions divided by a plurality of axes starting from the origin of the color difference signal space the color difference signal exists, An image conversion unit that converts the image IB using a conversion matrix that performs fc on the determined region, wherein the conversion matrix includes a coefficient that independently expands and contracts the plurality of axes, and a plurality of axes described in FJIJ. A color correction apparatus characterized in that the matrix is a matrix having a rotation angle and a rotation angle for independently rotating.

10. Input image signal and extract color difference signal with primary and secondary terms The color difference signal extraction unit that outputs

The color difference signal extracted by the color difference signal extraction unit is input, and it is determined in which region of the plurality of regions divided by the plurality of axes starting from the origin of the color signal space, the color difference signal is present, A first image conversion unit that converts the image signal using a first conversion matrix for the first-order term when the determined area is centered;

A second image conversion unit corresponding to the determined area and converting an image signal converted by the first image conversion unit using a second conversion matrix for the quadratic term. Prepare,

The first transformation matrix is a coefficient for the first-order term, and is a matrix having, as elements, a first-order coefficient for independently expanding and contracting the plurality of axes, and a rotation angle for independently rotating the plurality of axes. Wherein the second transformation matrix is a coefficient for a first-order term of HU, and expands and contracts the plurality of axes independently.- a second order coefficient, and a rotation angle for rotating the plurality of axes independently. A color correction device characterized in that the color correction device is a matrix having as elements.

11. A first step of separating an image signal into a luminance signal and a color difference signal by a luminance / chrominance signal separation unit;

The chrominance signals separated in the first step are input, and it is determined in which of the plurality of regions divided by the plurality of axes starting from the origin of the chrominance space the chrominance signal exists. A second step to determine,

A third step of converting the color difference signal by an axis conversion unit using a conversion matrix corresponding to the area determined in the second step;

A fourth step in which the color difference signal converted in the third step and the luminance signal separated in the first step are combined by a luminance / color difference signal combining unit.

Equipped with V

The transformation matrix includes a coefficient for expanding and contracting the plurality of axes independently, and A color correction method characterized in that it is a matrix having elements of a rotation angle for independently rotating the axis of the image.

12. A first step of separating an image signal into a luminance signal and a color difference signal by a luminance / chrominance signal separation unit;

A second step of inputting the color difference signal separated in the first step, and performing gamma conversion on the color difference signal by a gamma conversion unit based on a predetermined gamma force;

Input the color difference signal that has been gamma converted in the second step.

A third step of determining in which of a plurality of areas divided by a plurality of axes starting from the origin of the space the gamma-converted color difference signal is located; and A fourth step of converting the color difference signal by the axis conversion unit using a conversion matrix corresponding to the area determined in step 4), and the color difference signal converted in the fourth step and the first step. And a fifth step of combining the luminance signal separated by the step by a luminance / color difference signal combining unit.

The color correction method, wherein the conversion matrix is a matrix having, as elements, rotation angles for independently rotating the plurality of axes.

1 3. When the image signal of the specified point in the mouth image displayed on the screen is separated into a luminance signal and a color difference signal, this indicates where in the color difference signal space the color difference signal that is measured by α is located. A first step, and a second step of receiving a parameter-evening input for correcting the color difference signal obtained in the first step,

The parameters input in the second step of the FJIJ description are applied to the conversion matrix corresponding to the area corresponding to the location of the color difference signal obtained in the first step of H! J § The color difference signal obtained in the first step is converted to an axis conversion unit. And a fourth step indicating where the color difference signal converted in the third step exists in the color difference signal space.

The conversion matrix is a matrix whose elements are a coefficient for independently expanding and contracting a plurality of axes starting from the origin of the chrominance signal space and a rotation angle for independently rotating the plurality of axes. And a color correction display method.

14. A first signal indicating where in the color difference signal space the color difference signal obtained when the image signal at the designated point in the image displayed on the screen is separated into the luminance 15 and the color difference signal. Steps and

A second step of adding an axis passing from the origin of the FilJ color difference signal space to a position where the color difference signal exists in the color difference signal space;

,

U: a third step of receiving a parameter input for correcting the color difference signal obtained in the first step; and

,

Applying the parameters input in the third step of the RU to the conversion matrix corresponding to the area corresponding to the location of the color difference signal obtained in the first step, A fourth step of converting the color difference signal obtained in step 4 by the axis conversion unit, and indicating where in the color difference signal space the color difference signal converted in step 4 is located. With a fifth step,

The HIJ notation transformation matrix includes a coefficient for independently expanding and contracting the axis added in the second step and a rotation angle for independently rotating the axis added in the second step. A color-correction display method characterized by being a matrix with elements.

15. A first method for determining where in the color difference signal space the color difference signal obtained when an image signal at a designated point in the image displayed on the screen is separated into a luminance signal and a color difference signal. Steps and A second step of receiving a parameter input for correcting the color difference signal obtained in the first step;

f) Parameter U input in the second step U is applied to the conversion matrix centered on the area based on the location of the color difference signal determined in the first step of FJU Sd. -A third step in which the color difference signal obtained in the first step is converted by the axis conversion unit;

Mouth

A fourth step of displaying an image signal obtained by synthesizing the color difference signal converted in the third step and the luminance signal obtained in the first step on a screen,

The conversion matrix is a matrix having elements of: F) a coefficient for independently expanding and contracting a plurality of axes starting from the origin of the U gd color-difference signal space, and a rotation angle for independently rotating the plurality of axes described in FJIJ. A color correction display method characterized in that:

16. It is determined where in the color difference signal space the color signal obtained when the image signal of the specified occupation in the image displayed on the screen is separated into a luminance signal and a color difference signal. The first step and-

A second step of adding an axis passing from the origin of the color-difference signal space to a position where the color-difference signal exists in the color-difference signal space, and a color-difference signal obtained in the first step. A third step of accepting an input of a parameter for correcting the color difference signal, and the parameter inputted in the third step described above are converted into the color difference signal determined in the first step of the FJU Sli. A fourth step in which the color difference signal obtained in the first step is converted by an axis conversion unit by applying the conversion matrix to an area corresponding to the existence position of

a fifth step of displaying an image signal obtained by combining the color difference signal converted in the fourth step and the luminance signal obtained in the first step on a screen, The FJU transformation matrix is a coefficient for independently expanding and contracting the axis added in the second step, and a rotation angle for independently rotating the axis added in the second step in RIJ. Color correction display method characterized by being a matrix with

17. The first step of extracting the first color from the image signal by the color difference signal extraction unit,

色 Input the color difference signal extracted in the first step, and determine in which of the multiple regions divided by multiple axes the origin of the color difference signal space the color difference signal exists The second step,

A third matrix for converting the image signal by the image conversion unit using a conversion matrix corresponding to the area determined in the second step; and the third conversion matrix includes a plurality of axes. A color correction method characterized by the fact that it is a matrix with the coefficients for independently expanding and contracting and the rotation angles for independently rotating multiple axes.