JP2001313839A - Image-processing method and unit, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent gradation conversion processing and color correction processing from affecting each other when both processings are applied to the image data. SOLUTION: Image data S2 is obtained, by applying the gradation conversion processing to image data S0 through a gradation conversion means 22 and image data S3 is obtained by applying logarithmic conversion to the image data S2. A color correction amount setting means 32 sets the correction amount of lightness, saturation and hue of the specific color or the entire image, on the basis of the image data S3 subjected to the gradation conversion processing, when a color correction means 27 conducts color correction. Then the color correction means 27 conducts color correction on the basis of the correction amount.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image processing method and apparatus for performing gradation conversion processing and color correction processing on digital image data obtained by a digital camera or the like, and a program for causing a computer to execute the image processing method. The present invention relates to a recorded computer-readable recording medium.

[0002]

2. Description of the Related Art In a digital still camera (hereinafter referred to as a digital camera), an image obtained by photographing is recorded as digital image data on a recording medium such as an internal memory or an IC card provided inside the digital camera, and is recorded. Based on the digital image data, an image obtained by photographing can be displayed on a printer or a monitor. As described above, when an image obtained by a digital camera is printed, it is expected to have the same high-quality image quality as a photograph printed from a negative film.

In the field of printing, on the other hand, input image data is obtained by reading a color original with a scanner, and the input image data is subjected to desired image processing to generate output image data, which is hard-copyed to a printer. A system that outputs an image is used (for example, Japanese Patent Application Laid-Open No. H11-234523). This system converts input image data from RGB color signals to CMYK halftone
This is converted into a signal. First, a tone curve (gradation conversion table) and a color correction amount of the color correction unit are set in advance for the input image data, and the input image data is set based on the set tone curve and the color correction amount of the color correction unit. A three-dimensional lookup table (hereinafter referred to as 3DLUT) for converting data into output image data is created. Next, RG which is input image data
The B color signal is converted into a CMYK halftone% signal which is output image data by interpolating the 3DLUT. Printing is
The control is performed by controlling the amount of ink of each color by the dot% signal.

As an image processing method for performing gradation conversion processing and color correction processing on input image data, for example, a method described in Japanese Patent Application Laid-Open No. H10-210299 is known. This method uses a provisional saturation enhancement degree when obtaining an enlargement coefficient and a saturation enhancement degree for enhancing contrast based on input image data, that is, a gradation conversion parameter and a color correction processing parameter. In this case, the provisional saturation enhancement is changed according to the contrast expansion coefficient, and the color correction process is performed based on the changed saturation enhancement.

[0005]

However, in the above printing system, the parameters of the gradation conversion process (gradation conversion table) and the parameters of the color correction process (color correction amount) are both based on the input image data. Since the tone conversion process and the color correction process are performed on the input image data using these parameters, the tone conversion process becomes the color correction process, and the color correction process becomes the tone conversion process. Because of this, the gradation and color of the output image are different from those that should be originally obtained.

In the method described in Japanese Patent Application Laid-Open No. H10-210299, a temporary saturation enhancement degree, that is, a parameter of a color correction process is changed according to a parameter of a gradation conversion process. Is set based on the input image data, it is difficult to make the output image have the appropriate gradation and color, although the parameters of the gradation conversion processing are taken into account. Met.

The present invention has been made in view of the above circumstances, and when performing gradation conversion processing and color correction processing on image data, each processing does not affect each other and desired gradation and It is an object of the present invention to provide an image processing method and apparatus capable of obtaining a processed image having a color, and a computer-readable recording medium storing a program for causing a computer to execute the image processing method.

[0008]

According to the present invention, there is provided an image processing method for performing gradation conversion processing and color correction processing on image data to obtain processed image data. Performing the gradation conversion processing to obtain gradation-converted image data, setting a correction amount when performing the color correction processing based on the gradation-converted image data, and setting the gradation based on the correction amount. The color correction processing is performed on the converted image data to obtain the processed image data.

"Color correction processing" refers to processing for correcting the brightness, hue, and saturation of an image represented by gradation-converted image data. The “color correction processing” may be performed on the entire image represented by the gradation-converted image data, or may be performed only on a specific color such as flesh color or sky blue.

Here, the gradation conversion processing and the color correction processing in the present invention include not only the processing of directly performing the gradation conversion processing and the color correction processing on the image data, but also the processing of the image data and the processed image data. This includes creating a three-dimensional look-up table representing the correspondence and performing image processing by converting image data based on the three-dimensional look-up table.

An image processing apparatus according to the present invention is an image processing apparatus for performing gradation conversion processing and color correction processing on image data to obtain processed image data, wherein the gradation conversion processing is performed on the image data. Tone correction processing means for obtaining a tone-converted image data by performing the color correction processing based on the tone-converted image data; and a correction amount setting means for setting a correction amount when the color correction processing is performed. Color correction means for performing the color correction processing on the gradation-converted image data based on the image data to obtain the processed image data.

[0012] A computer-readable recording medium may be provided as a program for causing a computer to execute the image processing method according to the present invention.

[0013]

According to the present invention, a correction amount for performing a color correction process based on gradation-converted image data obtained by performing a gradation conversion process on image data is set. Therefore, it is possible to set a correction amount such that an appropriate color is obtained in consideration of the result of the gradation conversion process.
Therefore, by performing the color correction processing on the gradation-converted image data based on the correction amount, the gradation conversion processing and the color correction processing do not affect each other,
It is possible to obtain processed image data capable of reproducing an image having an appropriate gradation and color.

[0014]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic block diagram showing a configuration of an image output device to which an image processing device according to an embodiment of the present invention is applied. As shown in FIG. 1, an image output device 1 according to the present embodiment includes image data composed of color data R0, G0, and B0 from a memory card 2 storing image data S0 obtained by photographing a subject with a digital camera. Reading means 3 for reading S0; index image data S1 representing an index image by reducing image data S0;
1, setting information generating means 5 for analyzing the image data S0 and generating gradation setting information H0 necessary for setting a gradation conversion table T0 described later; A 3D LUT for performing gradation conversion processing and color correction processing on image data S0 when print output of S0 is performed, and 3DUT performing gradation conversion processing on index image data S11.
LUT creation means 6, monitor 7 for displaying index image data S11 'subjected to gradation conversion processing as an index image, input means 8 for making various inputs to 3DLUT creation means 6, and DCMY key for changing density 9 and
Using the 3DLUT created by the 3DLUT creating means 6, the image data S0 is converted to convert image data S12.
, A reducing means 11 for reducing the image data S0 when the number of pixels of the image data S0 is larger than the number of pixels for printing to obtain reduced image data S0 ', The converted image data S12 is enlarged when the number of pixels is smaller than
Enlarging means 12 for obtaining 2 ', sharpness processing means 13 for applying sharpness processing to converted image data S12 or enlarged image data S12' to obtain processed image data S13, and printing and outputting processed image data S13 A printer 14 for obtaining a print P.

The reading means 3 comprises a card reader for reading image data S0 from the memory card 2. Further, since the image data read from the memory card 2 is usually compressed, it is decompressed to obtain image data S0. The image data S0 includes information indicating the type of the digital camera that has taken the image (hereinafter referred to as camera type information).
Is provided as tag information, so that the camera type information is also read out at the same time. Here, as a standard for recording camera type information as tag information, for example, “Baseline TIF” which is adopted as an uncompressed file of an Exif file
F Rev.6.0RGB Full Color Image ".

The index image creating means 4 creates the index image data S11 by reducing the image data S0 by thinning or the like.

The setting information generating means 5 generates the setting information H0 as follows. Usually, in a digital camera, an auto exposure control process (AE process) and an auto white balance adjustment process (AWB process) are performed on the assumption that the image data S0 is reproduced on a monitor. However, when the image data S0 is reproduced in the printer, the AE processing and the AWB processing (hereinafter, referred to as AE / AWB processing) performed in the digital camera are not enough.
It is necessary to perform E / AWB processing. Setting information generating means 5
Is for estimating a correction amount necessary for performing an AE / AWB process optimum for printing for each of the RGB color signals constituting the image data S0, and including this correction amount in the setting information H0. Therefore, for example, as described in JP-A-11-220609, an average value is obtained for each of the R, G, and B signals constituting the image data S0, and a correction value is set so that the average value becomes a target value suitable for printing. Then, this correction value is included in the setting information H0 as a correction amount and output. This correction amount is used to correct both the exposure amount and the white balance.

In the setting information generating means 5, as will be described later, a 3DLUT
Is generated, a correction amount for nonlinearly correcting gradation highlights and shadows is obtained, and this correction amount is also included in the setting information H0. Here, the printer has a narrow reproduction range of density, jumps to a highlight portion of an image, and is in a state in which a shadow portion is easily collapsed. For this reason, the setting information generating means 5 may use the method described in, for example, JP-A-11-331596 to harden the gradation on the highlight side when the print density is increased by AE processing or AWB processing. When the print density is reduced, the tone on the shadow side is softened.On the other hand, if the print density decreases, the correction amount is calculated so that the density on the highlight side is softened and the tone on the shadow side is hardened. Are included in the setting information H0.

Further, in the setting information generating means 5, the tag information of the image data S0 is read, and the camera type information of the tag information is included in the setting information H0. If the tag information includes strobe information, this is also included in the setting information H0.

The monitor 7 displays the index image data S1
The index image represented by 1 'is displayed. Further, at the time of correcting a gradation curve described later, a gradation curve is displayed together with the index image. In this embodiment, six index images are displayed simultaneously.

The input means 8 comprises a keyboard, a mouse, etc. for making various inputs to the 3DLUT creating means 6. Here, from the input means 8, a type of a gradation (hereinafter referred to as a reference gradation) which is a reference when the 3DLUT is created is input. Note that the reference gradation refers to a print P having an appropriate gradation when printing is performed by the printer 14.
Represents the gradation at which the gradation conversion process is performed on the image data so as to obtain. Here, as the reference gradation, for example, a standard gradation, a gradation for cloudy weather, a gradation for backlight, and a gradation for a close strobe scene can be selected. By inputting the reference gradation, a gradation curve representing the selected reference gradation is set to 3DLU.
This is set in the T creating means 6. In some cases, it is necessary to modify the gradation curve so that a desired gradation can be obtained. In such a case, display the gradation curve on the monitor 7 and correct the gradation curve using the input means 8. Can be.

The DCMY key 9 comprises four keys for correcting the densities of the colors D, C (cyan), M (magenta), and Y (yellow) of the entire image. In the 3DLUT creating means 6, the density of the entire image and each color is changed. The correction of the gradation curve input from the input means 8 and the DCMY key 9
The change of the density input from the
Is reflected in the index image displayed in.

The 3DLUT creating means 6 creates a 3DLUT as follows. FIG. 2 is a schematic block diagram showing the configuration of the 3DLUT creating means 6. The image data S
If 0 is 8-bit data for each color of RGB, 256-bit data is converted to create a 3DLUT for converting all data.
³ data is required, and it takes a long time to create a 3DLUT. Accordingly, in the present embodiment, it is assumed to create each color data R0, G0, 0,7,15 by reducing the B0 number of bits, ... 33 ³ 3DLUT comprising data for each color 33 of 247,255.

As shown in FIG. 2, the 3DLUT creating means 6
Is ITU-R BT. 709 (REC. 709), the true color data R0 ', G0', and B0 'are calculated based on the following equations (1) to (3). Logarithmic conversion means 21 for obtaining the image data S1 by logarithmically converting the image data, and gradation conversion means 22 for performing a process of converting the grayscale of the logarithmically converted image data S1 to obtain image data S2; A gradation setting unit 23 for setting a gradation conversion table T0 used for gradation conversion in the gradation conversion unit 22, a memory 24 storing a plurality of gradation curves, and color data obtained by performing inverse logarithmic conversion of the image data S2. Antilogarithmic conversion means 25 for obtaining image data S3 composed of R3, G3, B3; and color data R3, G3, B3 constituting image data S3 are converted to data L3, C3 representing lightness L ^* , saturation C ^* and hue HA. , H3 Conversion to LCH
A conversion unit 26; a color correction unit 27 that performs color correction processing on the data L3, C3, and H3 to obtain color correction data L4, C4, and H4; and a color correction data L4, C4, and H4.
Is converted to an sRGB color space, which is a monitor color space, and color corrected image data S4 composed of color data R4, G4, and B4.
SRGB conversion means 28 for obtaining color correction image data S4
To a printer color space to obtain printer image data S5, and LUT creation means 30 for creating a 3DLUT based on the printer image data S5 and the image data S0. The color correction means 27 has a memory 31 storing a plurality of color correction menus.
And image data S3 based on the data L3, C3 and H3.
Is connected to a color correction amount setting unit 32 that sets a correction amount for correcting a specific color (skin color in the present embodiment) included in the image represented by.

Pr = R0 / 255 Pg = G0 / 255 (1) Pb = B0 / 255 R0 ′ = ((Pr + 0.099) /1.099) ^2.222 G0 ′ = ((Pg + 0.099) /1.099) ^2.222 (Pr, (Pg, Pb ≧ 0.081) (2) B0 ′ = ((Pb + 0.099) /1.099) ^2.222 R0 ′ = Pr / 4.5 G0 ′ = Pg / 4.5 (Pr, Pg, Pb <0.081) (3) B0 '= Pb / 4.5 The memory 24 has a reference gradation curve including a standard gradation curve, a gradation curve for cloudy weather, a gradation curve for backlight, and a gradation curve for a close-up strobe scene. In addition, a plurality of gradation curves corresponding to camera types are stored.

In the gradation setting means 23, a gradation conversion table T0 for performing gradation conversion of the image data S1 is set as follows. FIG. 3 is a diagram for explaining the setting of the gradation conversion table T0. The gradation conversion table T0 includes color data R1 and G constituting the image data S1.
1, B1 from the first quadrant to the fourth quadrant, and performs gradation conversion to form color data R2, G2, B2 constituting image data S2.
Is what you get. In the gradation setting means 23, a gradation conversion table T0 is set for each of the RGB colors. First, the setting information H0 is input to the gradation setting means 23, and based on the camera type information in the setting information H0,
A gradation curve corresponding to the camera type information is read from the memory 24. On the other hand, a default standard gradation curve is read out from the memory 24 as a reference gradation curve. Is read, and if it is input that the backlight gradation curve is to be read, the backlight gradation curve is read, and if the reading of the proximity strobe gradation curve has been input. Reads out a gradation curve for a proximity strobe.

The gradation curve C1 for each camera type is set in the first quadrant as shown in FIG. Here, in a digital camera, the image quality of a reproduced image differs depending on the type of camera, such as the manufacturer and model of the digital camera. Therefore, the gradation curve C1 is created according to the camera type so as to absorb the gradation characteristics of each camera in order to obtain an image of constant quality regardless of the camera type. When the color data R1, G1, and B1 are converted by the gradation curve C1, data representing a logarithmic exposure amount in which the gradation characteristics in the camera are compensated is obtained.

In the second quadrant, a straight line C2 for correcting the exposure amount is set. Although the straight line C2 for correcting the exposure amount is basically a straight line passing through the origin, the straight line C2 is moved in the direction of arrow A based on the exposure amount included in the setting information H0 and the correction amount for correcting the white balance. The amount of exposure is corrected by the parallel movement. Then, AE / AWB processing suitable for printing is performed on the straight line C2, and data representing the reflection density of the actual subject is obtained.

In the third quadrant, a reference gradation curve is set. Here, it is assumed that a standard gradation curve C3 is set. The standard gradation curve C3 is an S-shaped curve, and the middle portion corresponds to γ = 1.6. Here, in the present embodiment, the conversion based on the gradation curve C3 is referred to as γ conversion. Then, density data suitable for printing can be obtained from the gradation curve C3.

In the fourth quadrant, a gradation curve C4 for nonlinearly correcting a highlight portion and a shadow portion of an image is set. The correction amount of the gradation curve C4 is determined according to the correction amount of the highlight portion and the shadow portion included in the setting information H0. Then, the image data S is obtained by the gradation curve C4.
2, color data R2, G2, and B2 can be obtained.

The gradation conversion table T0 is changed according to the input of the input means 8 and / or the DCMY key 9. Here, when the DCMY key 9 is pressed, C, M, and Y of the index image displayed on the monitor 7 are shifted. Here, the shift amounts of C, M, and Y are set to R, G, and B.
This is to change the gradation conversion table T0 by converting to a density shift amount. That is, the shift amounts of the R, G, and B densities according to the number of times the DCMY key 9 is pressed are set in advance, and the R, G, and B density values are set according to the number of times the DCMY key 9 is pressed.
The concentrations of G and B are changed. Specifically, the density of R, G, and B is changed by moving the straight line C2 in the second quadrant in the direction of arrow A in accordance with the number of times the DCMY key 9 is pressed. Further, depending on the input from the input means 8, the gradation curve C1 in the first quadrant or the gradation curve C3 in the third quadrant may be used.
Is changed. In this case, the gradation curves C1 and C3 for each color are displayed on the monitor 7 together with the index image, and the input means 8 is used to observe the index image so that the gradation curves C1 and C3 can be obtained as desired by the user.
What is necessary is just to change C3. Then, as described above, the gradation curve C
1. The gradation conversion table T0 is changed by changing the straight line C2 and / or the gradation curve C3.

The gradation conversion means 22 converts the image data S1 according to the gradation conversion table T0 set by the gradation setting means 23 to obtain image data S2.

The logarithmic conversion means 21 and the gradation conversion means 2
2, and the antilogarithmic conversion unit 25 performs all the processes in the RGB color space.

The LCH conversion means 26 converts the image data S3 into R
While converting from the GB color space to the L ^* a ^* b ^* color space,
Data L3, C3, and H3 representing lightness L ^* , chroma (chroma value) C ^*, and hue angle HA are obtained. Hereinafter, this conversion will be described. Image data S0 acquired by the digital camera is based on ITU-R BT. 709 (R
EC. 709), the color data R3, G constituting the image data S3 based on the following equation (4).
3, B3 are converted to CIE1931 tristimulus values X, Y, Z.

X R3 Y = | A | · G3 (4) Z B3 Here, the matrix | A | is the color data R3, G3, B
3 is a matrix for converting 3 into tristimulus values X, Y, and Z. For example, the following values can be used.

0.4124 0.3576 0.1805 | A | = 0.2126 0.7152 0.0722 (5) 0.0193 0.1192 1.0571 Instead of the matrix | A |, the tristimulus values X, Y, and Z may be obtained by a look-up table.

Next, from the tristimulus values X, Y, and Z, CIE1976L ^* (= L3) by the following equations (6) to (8):
The chroma value C ^* (= C3) and the hue angle HA (= H3) are obtained.

A ^* = 500 {f (X / Xn) -f (Y / Yn)} b ^* = 200 {f (Y / Yn) -f (Z / Zn)} (6) L ^* = 116 (Y / Yn) ^1/3 -16 (when Y / Yn> 0.008856) L ^* = 903.25 (Y / Yn) (when Y / Yn ≦ 0.008856) where X / Xn, Y / Yn, Z / When Zn> 0.008856 f (a / an) = (a / an) ^1/3 (a = X, Y, Z) When X / Xn, Y / Yn, Z / Zn ≦ 0.008856 f (a / an) = 7.787 (a / an) +16/11
6 Xn, Yn, and Zn are tristimulus values for white, and can be substituted by tristimulus values corresponding to CIE-D65 (a light source with a color temperature of 6500K).

C ^* = (a ^{* 2} + b ^{* 2} ) ^1/2 (7) HA = tan ⁻¹ (b ^* / a ^* ) (8) The color correction means 27 includes R, G, B, C, M , Y, YellowGr
een (YG), BlueSky (BS), highlight color S
The lightness, saturation, and hue of 11 colors of K (HL), the skin color SK (MD) of the intermediate density, and the skin color SK (SD) on the shadow side are corrected. Specifically, the following expressions (9) to
As shown in (11), the data L3, C3, and H3 are corrected to obtain corrected data L4, C4, and H4.

[0041]

(Equation 1) Here, i: R, G, B, C, M, Y, SK, BS j: SK (HL), SK (MD), SK (SD) LPi, LPj: Lightness change degree CPi, CPj: Saturation change degree HPi, HPj: Hue change degree Wi, Wj: Intensity function Δl: Lightness change due to gradation change Δc: Saturation change due to gradation change Δh: Hue change due to gradation change Lightness change degrees LPi, LPj , Saturation change degree CPi, CPj
The hue change degrees HPi and HPj are set by a color correction menu stored in the memory 31. FIG. 4 shows a color correction menu. Here, the memory 31 has a reference color correction menu for performing common color correction on all the data L3, C3, and H3, and a model color correction menu for performing color correction according to the camera type. It is remembered. When the setting information H0 is input to the color correction means 27, a model color correction menu corresponding to the camera type is read from the memory 31 based on the camera type information included in the setting information H0. On the other hand, a default standard color correction menu is stored in the memory 3 as a reference color correction menu.
However, if the input means 8 inputs that the color correction menu for cloudy weather is read, the color correction menu for cloudy weather is read, and the color correction menu for backlight is input. In this case, the backlight correction color correction menu is read, and when the instruction to read the close strobe color correction menu is input, the close strobe color correction menu is read. Here, in the color correction menu, numerical values indicating how much the lightness, saturation, and hue should be corrected are set, and the color correction unit 27 sets the numerical values set in the reference color correction menu and the model color correction menu, and According to the correction amount set by the color correction amount setting means 32 as described later, Expression (9)
The brightness change degrees LPi and LPj, the saturation change degrees CPi and CPj, and the hue change degrees HPi and HPj in (11) to (11) are set. The degree of change for each color is obtained as the sum of the reference color correction menu, the model color correction menu, and the numerical value of the correction amount set by the color correction amount setting means 32.

The intensity function Wi is determined by the following equation (12).

Wi = F (d) d = √ ((Li−L) ² + (ai−a) ² + (bi−b) ² ) (12) where Li, ai, and bi are R, G, B, C, M, Y,
This is the center color in the L ^* a ^* b ^* color space of YG and BS,
R, G, B, C, M, and Y are colorimetric values of each color of Macbeth Color Checker (registered trademark; manufactured by Macbeth A division kollmorgen), and YG and BS are image data S0. Are the average colorimetric values of the green leaves and the sky portion of the image represented by Also, d is L ^* between the center colors Li, ai, bi and the values of L ^* , a ^* , b ^* obtained in the LCH conversion means 26 ^.
A distance in the a ^* b ^* color space, and F (d) has a constant value until the distance d reaches a predetermined value (here, 30) as shown in FIG. 5, for example. This is a function such that the value becomes smaller as becomes larger.

On the other hand, the intensity function Wj is an intensity function for flesh color, and L ^* a ^* b ^{* of the} image represented by the image data S0 ^.
The skin color SK (HL) on the highlight side in the color space, the skin color SK (MD) of the intermediate density, and the skin color SK on the shadow side
The statistical distribution range of (SD) is obtained, and in the distribution, as shown in FIG. 6, the value of the peripheral portion is set to be small and the value of the central portion is set to be large (however, 0 ≦ Wj ≦ 1).

In one of the index images displayed on the monitor 7 as shown in FIG.
G, B, C, M, Y, YG, BS, SK (HL), SK
An arbitrary color other than (MD) and SK (SD) is designated, the degree of change of the color is set with the designated color as the center color, and the change of the color is reflected in the above equations (9) to (11). You may. In this case, if points A and B in FIG. 7 are designated,
A color in a 5 × 5 range around the points A and B is obtained, and a color correction menu is set for the color as shown in FIG. 8, and the correction data L4 and
C4 and H4 are required.

.DELTA.l, .DELTA.c and .DELTA.h are changes in lightness, saturation and hue of flesh color due to the non-linear gradation conversion set in the fourth quadrant of the gradation setting means 23, and are obtained as follows. Can be That is, the color data R1, G1,
B1 and the color data R2, G2, and B2 after the gradation conversion are subjected to the processing of the above equations (4) to (8) and the processing in the antilogarithmic conversion means 25, and the brightness L ^* , the chroma The change amounts ΔL ^* , ΔC ^*, and ΔHA of the value C ^* and the hue angle HA are calculated. Then, the following equations (13) to (1)
As shown in 5), Δl, Δc, Δh can be obtained by multiplying the variation amounts ΔL ^* , ΔC ^*, and ΔHA by the skin color intensity function Wj shown in FIG.

Δl = ΔL ^* × Wj (13) Δc = ΔC ^* × Wj (14) Δh = ΔHA × Wj (15) The color correction amount setting means 32 corrects the brightness, saturation and hue as follows. Set. The color correction amount setting means 32 receives image data S3 subjected to gradation processing and data L3, C3, and H3 representing lightness, saturation, and hue of an image represented by the image data S3.
3 from the image represented by, for example, Japanese Patent No. 2762221.
The skin color area is extracted by the method described in the reference number. In the method of Japanese Patent No. 2763221, a color area having the same or similar hue is obtained on a color image, a circle or an ellipse inscribed at a boundary portion of the obtained color area is assumed, and a circle or a long area having a large area is obtained. The color area is divided in order from a circle, and at least one of the divided areas is extracted as a skin color area of a person's face. Then, the average value of the data L3, C3, and H3 corresponding to the extracted flesh color region, that is, the average values La, Ca, and Ha of the brightness, saturation, and hue of the extracted flesh color region are obtained. When a plurality of skin color regions are extracted, the average values La, Ca, Ha for each region are extracted.
Is required.

Here, in the color correction amount setting means 32, as shown in the following equation (16), three preferred levels of skin color brightness, saturation and hue are set according to the brightness of the skin color area. I have.

When 80 <La, Lh, Ch, Hh When 40 <La ≦ 80, Lm, Cm, Hm (16) When La ≦ 40, Ls, Cs, Hs where Lh, Ch, Hh are bright Preferred lightness, saturation and hue of the skin, Lm, Cm and Hm represent the lightness, saturation and hue of the preferred medium skin, and Ls, Cs and Hs represent the brightness, saturation and hue of the preferred dark skin, respectively. .

Then, according to the brightness of the extracted flesh-colored area, the average values La, Ca, Ha and Lh, Ch, H of the flesh-colored area are calculated as shown in the following equations (17) to (19).
h, Lm, Cm, Hm or a difference from Ls, Cs, Hs, and a coefficient β (0 ≦ β ≦
1), and the correction amounts ΔLh, ΔCh, ΔHh, ΔLm, Δ of the lightness, saturation, and hue according to the brightness of the skin color region
Cm, ΔHm, ΔLs, ΔCs, and ΔHs are obtained.

ΔLh = β (Lh−La) ΔCh = β (Ch−Ca) (17) ΔHh = β (Hh−Ha) ΔLm = β (Lm−La) ΔCm = β (Cm−Ca) (18) ΔHm = Β (Hm−Ha) ΔLs = β (Ls−La) ΔCs = β (Cs−Ca) (19) ΔHs = β (Hs−Ha) Note that the average values La, Ca, and Ha are bright skin, medium brightness. A membership function (affiliation function) for determining the degree of affiliation indicating how much the skin color and the dark skin belong to each other is set, and the average value La, C of each skin color area is set.
a, b, and c (where a, b, and c) belong to the bright skin, the medium-brightness skin, and the dark skin in each skin color area based on a and Ha.
+ B + c = 1), and based on the affiliations a, b, and c, as shown in the following equations (20) to (22), the correction amounts of the brightness, saturation, and hue corresponding to the brightness of the skin color area ΔL
h, ΔCh, ΔHh, ΔLm, ΔCm, ΔHm, ΔL
s, ΔCs, and ΔHs may be obtained.

ΔLh = aβ (Lh−La) ΔCh = aβ (Ch−Ca) (20) ΔHh = aβ (Hh−Ha) ΔLm = bβ (Lm−La) ΔCm = bβ (Cm−Ca) (21) ΔHm = Bβ (Hm−Ha) ΔLs = cβ (Ls−La) ΔCs = cβ (Cs−Ca) (22) ΔHs = cβ (Hs−Ha) The sRGB conversion means 28 corrects the correction data L4, C4, H4.
By solving the above equations (7) and (8) in reverse, the corrected a ^* and b ^* are obtained. For the corrected a ^* , b ^* and L ^* , the equation (6) is reversed. By solving, the corrected tristimulus values X5, Y5, Z5 are obtained. Then, the tristimulus values X5, Y5, Z5 are converted into color data R4 ', G4', B4 'by the following equation (23).

R4 ′ X5 G4 ′ = | A | ⁻¹ · Y5 (23) B4 ′ Z5 Further, the color data R4, G4,
B4 is obtained and used as color correction image data S4 in the sRGB color space for display on the monitor 7.

R4 = 255 × ((R4 ′ + 0.055) /1.055) ^2.4 G4 = 255 × ((G4 ′ + 0.055) /1.055) ^2.4 (0.03929 ≦ R4 ′, G4 ′, B4 ′ ≦ 1) B4 = 255 × ((B4 ′ + 0.055) /1.055) ^2.4 R4 = 255 × R4 ′ / 12.92 G4 = 255 × G4 ′ / 12.92 (0 ≦ R4 ′, G4 ′, B4 ′ <0.03929) B4 = 255 × B4 '/12.92 (24) The printer conversion unit 29 converts the color correction image data S4 in the sRGB color space into a color space for printing.
To convert the color correction image data S4 to obtain printer image data S5.

The LUT creating means 30 obtains, for each color, the correspondence between the color data R0, G0, B0 constituting the image data S0 and the color data R5, G5, B5 constituting the printing image data S5. and 33 ³ of the three-dimensional look-up table (3DLUT).

The 3DLUT creating means 6 receives the index image data S11 and performs a gradation conversion process. The gradation conversion processing is performed on the index image data S11 by the gradation conversion means 22 without reducing the number of bits. Only the tone conversion process using the tone conversion table T0 is performed, the color image is converted into the sRGB color space without being subjected to the color correction process in the color correction unit 27, and the index image data S11 subjected to the tone conversion process. '. At this time, the index image data S11 is 3DLU
Since the tone conversion table T0 is not used to create the T, the tone setting means 23 uses the tone conversion table T0 which is sequentially set by reflecting the density shift due to the depression of the DCMY key 9 or the change of the tone curve. Are sequentially subjected to gradation conversion and output as index image data S11 '. Thereby, the index image whose gradation has been changed can be displayed on the monitor 7 in real time.

Returning to FIG. 1, the 3DLUT created by the 3DLUT creating means 6 is input to the processing means 10.
Then, the image data S0 is converted by the 3DLUT to obtain converted image data S12. At this time, since the 3DLUT is created by 33 ^third data, the color data constituting the converted image data S12,, for example, JP-A-2-8
As described in No. 7192, the 3DLUT is obtained by volume interpolation or area interpolation.

The number of pixels of the digital camera that has acquired the image data S0 is various, and some of them have a pixel number less than the number required for printing or a number larger than the number required for printing. For this reason, when the image data S0 has the number of pixels equal to or larger than the number of pixels necessary for printing, the reduction unit 11 reduces the image data S0 at the previous stage of the processing unit 10 to obtain reduced image data S0 ', and the reduced image data S0 Is converted by the 3DLUT to obtain converted image data S12. On the other hand, if the image data S0 is less than the number of pixels required for printing,
In the subsequent stage, the converted image data S12 obtained by the processing means 10 is enlarged by the enlargement means 12 to obtain enlarged image data S12 '.

The sharpness processing means 13 performs a sharpness process on the converted image data S12 or the enlarged image data S12 'according to, for example, the following equation (25) to obtain processed image data S13. In equation (25), the converted image data S12 is subjected to sharpness processing.

S13 = S12 + γ (S12−S12us) (25) where S12us: the blurred image data of the converted image data S12 γ: the emphasis degree The emphasis degree γ is set to the reduction ratio by the reduction unit 11 or the enlargement ratio by the enlargement unit 12. It may be changed accordingly.

Next, the operation of this embodiment will be described. FIG. 9 is a flowchart showing the operation of the present embodiment. First, the image data S0 is read by the reading means 3 from the memory card 2 in which the image data S0 obtained by photographing with the digital camera is stored (step S1). In the index image creating means 4, index image data S11 representing an index image of the image data S0 is created (step S2),
It is input to the 3DLUT creation means 6. On the other hand, the setting information generation means 5 generates the setting information H0 (step S3) and inputs it to the 3DLUT generation means 6.

The gradation setting means 23 of the 3DLUT creating means 6
, A gradation conversion table T0 for converting the image data S0 is set based on the setting information H0 (step S4). The data S11 is subjected to gradation conversion (step S5), and an index image is displayed on the monitor 7 without performing color correction (step S6). The user observes the index image, and if necessary, inputs data 8 or DCMY key 9.
(Step S7), the gradation and / or density of the index image is corrected (step S7).
8). Then, returning to step S4, the tone conversion table T0 is newly set based on the corrected tone and / or density, and the index image data S11 is tone-converted by the newly set tone conversion table T0. The processing from step S4 to step S7 displayed on the monitor 7 is repeated. If there is no correction or the correction has been completed, step S7 is denied, and gradation conversion is performed on the image data S0 by the gradation conversion table T0 finally set (step S9), and color correction is further performed. Is performed (step S10). Further, conversion into the sRGB color space and conversion into the print color space are performed (step S11), and print image data S5 is obtained. Then, the LUT creation means 30 finds the correspondence between the image data S0 and the print image data S5 for each color of RGB, and creates a 3DLUT (step S12).
The process ends.

The image data S0 read from the memory card 2 is processed by the 3DLUT by the processing unit 10.
Are subjected to reduction processing by the reduction means 11 and enlargement processing by the enlargement means 12 if necessary, further subjected to sharpness processing by the sharpness processing means 13, and output as a print P by the printer 14.

As described above, in the present embodiment, the correction amount for performing the color correction process based on the image data S3 after the gradation conversion process is performed on the image data S0 is set. It is possible to set a correction amount such that an appropriate color is obtained in consideration of the result of the gradation conversion processing.
Therefore, the 3DLU set based on this correction amount
If the gradation conversion processing and the color correction processing are performed on the image data S0 by T, the image having the appropriate gradation and color can be reproduced without the gradation conversion processing and the color correction processing affecting each other. The processed image data can be obtained.

In the above-described embodiment, the correction amount for the skin color is set by the color correction amount setting means 32. However, other colors such as R, G, B, C, M, Y, YG, and BS are set. May be set. In this case, a region having each color is extracted from the image represented by the image data S3, and an average value of the region is determined. As shown in the above equations (17) to (19), this average value is determined in advance. The correction amount may be obtained based on the difference from the preferred color. Further, the correction amount for the entire image may be set. In this case, an average value of the brightness, saturation, and hue of the entire image may be obtained, and the correction amount may be obtained based on a difference between the average value and predetermined preferable brightness, saturation, and hue.

In the above embodiment, the image processing apparatus according to the present invention is used for creating a 3DLUT.
The gradation conversion processing and the color correction processing may be performed in real time. Although the image processing apparatus according to the present invention is applied to an image output apparatus, the image processing apparatus may be used alone.

Further, in the above embodiment, the gradation conversion processing and the color correction processing are performed on the image data obtained by the digital camera, but the original or the film on which the image is recorded is read by the scanner. Gradation conversion processing and color correction processing can be performed on the obtained image data in the same manner as described above.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram illustrating a configuration of an image output device to which an image processing device according to an embodiment of the present invention is applied.

FIG. 2 is a schematic block diagram showing a configuration of a 3DLUT creating unit.

FIG. 3 is a diagram for explaining setting of a gradation conversion table;

FIG. 4 is a diagram showing a color correction menu.

FIG. 5 is a diagram showing an example of an intensity function.

FIG. 6 is a diagram showing an example of an intensity function for skin color.

FIG. 7 is a diagram showing one of the index images displayed on the monitor.

FIG. 8 shows an additional color correction menu.

FIG. 9 is a flowchart showing the operation of the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image output device 2 Memory card 3 Reading means 4 Index image creating means 5 Setting information generating means 6 3DLUT creating means 7 Monitor 8 Input means 9 DCMY key 10 Processing means 11 Reduction means 12 Enlarging means 13 Sharpness processing means 14 Printer 21 Logarithmic conversion Means 22 Gradation conversion means 23 Gradation setting means 24, 31 Memory 25 Antilogarithmic conversion means 26 LCH conversion means 27 Color correction means 28 sRGB conversion means 29 Printer conversion means 30 LUT generation means 31 Memory 32 Color correction amount setting means

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (Reference) H04N 1/46 H04N 1/46 Z 5C079 9/79 9/79 HF term (Reference) 2C087 AA15 AA16 BB10 BD31 BD35 BD36 2C262 AA24 AB11 AC08 BA02 BA09 BC01 BC10 BC13 EA02 FA14 FA20 5B057 CA01 CB01 CC03 CE11 CE17 5C055 AA05 AA06 BA06 BA08 EA05 EA06 FA22 GA00 HA17 HA37 5C077 LL19 MP08 PP15 PP32 PP33 PP35 PP37 P01B03 H35 PP01

Claims (3)

[Claims] 1. An image processing method for performing a gradation conversion process and a color correction process on image data to obtain processed image data, wherein the gradation conversion process is performed on the image data. Obtaining image data, setting a correction amount when performing the color correction process based on the gradation-converted image data, and performing the color correction process on the gradation-converted image data based on the correction amount. Image processing method to obtain the processed image data. 2. An image processing apparatus for performing gradation conversion processing and color correction processing on image data to obtain processed image data, wherein the image data is subjected to the gradation conversion processing and gradation conversion is performed. Gradation conversion processing means for obtaining image data; correction amount setting means for setting a correction amount when performing the color correction processing based on the gradation-converted image data; and gradation conversion based on the correction amount. An image processing apparatus comprising: a color correction unit that performs the color correction process on the processed image data to obtain the processed image data. 3. A computer-readable recording medium on which a program for causing a computer to execute an image processing method of performing a gradation conversion process and a color correction process on image data to obtain processed image data is recorded. A program for performing the gradation conversion processing on the image data to obtain gradation-converted image data; and setting a correction amount when performing the color correction processing based on the gradation-converted image data. Performing the color correction process on the gradation-converted image data based on the correction amount to obtain the processed image data.