JP2010251940A - Image processor, image processing method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently prepare a profile that gives a correspondence relation between a colorimetric value and image data. <P>SOLUTION: A simple lattice 3D-LUT creating part 84 creates a lookup table defining a correspondence relation between an input color space and a target color space on the basis of a correspondence relation between the colorimetric value of a plurality of color patches and an input value obtained by inputting the color patches by an image input device. A lightness function creating part 85 creates a lightness function for converting lightness of the input value defined in the lookup table so as to minimize a color difference between itself and the colorimetric value. A profile creating part 78 creates a profile on the basis of the lookup table with the lightness function applied therein. The lightness function creating part 85 determines a knee point in the lightness function on the basis of a knee angle of the average lightness function of the lightness functions that have been already created and the output value of the knee point. This image processor is applicable to a personal computer. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing device, an image processing method, and a program.

  An ICC (International Color Consortium) standard that, when a color input by a computer input device is output by an output device, the output color is reproduced as the same color as the input color regardless of the device-specific characteristics Adoption is progressing. In this ICC standard, profile data giving a correspondence relationship between the colorimetric values of the color chart and image data is given to each input / output device, and the image data of each input / output device is used as a reference measurement via this profile. Coloring independent of each device is performed by being associated with the color value.

  For example, a method has been proposed in which a device-dependent color space used in a digital still camera is converted into an independent color space and a profile for performing color correction for each scene is created (see Patent Document 1).

JP 2007-258868 A

  However, in creating a profile for converting a device-dependent color space into an independent color space, the necessary parameters (for example, brightness function) are usually determined sensuously by the user based on previous experience. . As a result, it may take time to obtain a desired image, and variations may be produced in the finished image.

  The present invention has been made in view of such a situation, and makes it possible to effectively generate a profile.

  An image processing apparatus according to an aspect of the present invention provides a colorimetric value of a plurality of color patches in an image processing apparatus that creates a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space. And a lookup table creating means for creating a lookup table that defines the correspondence between the input color space and the target color space based on the correspondence between the color patch and the input value input by the image input device, and the lookup table Profile based on a brightness function creation means for creating a brightness function for converting the brightness of the input value specified in to a colorimetric value so as to minimize the color difference, and a lookup table to which the brightness function is applied And a brightness function creating means that calculates a knee point in the brightness function as a result of the brightness function already created. The look-up table creation means determines the correspondence between the input color space and the target color space based on the correspondence between the colorimetric value and the input value. Create a specified matrix, create a single-grid three-dimensional lookup table that stores the correspondence between the input color space specified by the matrix and the target color space, and use the matrix to determine the white point of the input color space Based on the white brightness obtained by conversion, the brightness of each color value of the target color space defined in the single grid three-dimensional lookup table is normalized, and the single grid three-dimensional lookup table is applied to the multi-grid lookup table. And applying a brightness function to the multi-grid lookup table.

  For the weighting coefficient whose hue and saturation are corrected according to the magnitude, the weighting function creating means for creating a weighting coefficient having a magnitude inversely proportional to the magnitude of the output value in the nonlinear part of the brightness function, The profile creation unit can create a profile based on a lookup table to which the weighting factor created by the weighting factor creation unit is applied.

  Display control means for displaying an operation screen having an input unit for receiving a matrix, lightness function, or weighting factor output instruction, and parameter output for outputting a matrix, a lightness function, or a weighting factor according to the instruction received by the input unit Means.

  The look-up table creation means can avoid applying the brightness function.

  An image processing method or program according to an aspect of the present invention provides a colorimetric measurement of a plurality of color patches when creating a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space. A lookup table creation step for creating a lookup table for defining the correspondence between the input color space and the target color space based on the correspondence between the value and the input value obtained by inputting the color patch by the image input device; Based on the brightness function creation step that creates a brightness function for converting the brightness of the input value specified in the table so that the color difference from the colorimetric value is minimized, and a lookup table to which the brightness function is applied, A profile creation step for creating a profile, wherein the brightness function creation step calculates a knee point in the brightness function, Based on the average brightness function knee angle and the output value of the knee point of the brightness function created in step 2, the lookup table creation step determines the input color space based on the correspondence between the colorimetric value and the input value. Create a matrix that defines the correspondence with the target color space, create a single-grid three-dimensional lookup table that stores the correspondence between the input color space defined by the matrix and the target color space, and use the matrix Based on the white brightness obtained by converting the white point of the input color space, the brightness of each color value in the target color space defined in the single grid three-dimensional lookup table is normalized, and the single grid three-dimensional lookup table is A multi-grid look-up table is created by application, and a brightness function is applied to the multi-grid look-up table.

  In the image processing apparatus, the image processing method, or the program according to one aspect of the present invention, the input color space is based on the correspondence between the colorimetric values of the plurality of color patches and the input values obtained by inputting the color patches by the image input device. A lookup table that defines the correspondence between the target color space and the target color space is created, and a brightness function is created to convert the brightness of the input value defined in the lookup table so that the color difference from the colorimetric value is minimized. A profile is created based on the lookup table to which the lightness function is applied, and in the lightness function creation, the knee point in the lightness function is the average lightness function knee angle and knee point output of the lightness function already created The input color space and the purpose based on the correspondence between the colorimetric value and the input value. A matrix that defines the correspondence with the space is created, and a single-grid three-dimensional lookup table that stores the correspondence between the input color space defined by the matrix and the target color space is created and input using the matrix The brightness of each color value in the target color space specified in the single grid 3D lookup table is normalized based on the white brightness obtained by converting the white point of the color space, and the single grid 3D lookup table is applied. Thus, a multi-grid look-up table is created, and a brightness function is applied to the multi-grid look-up table.

  According to the present invention, it is possible to efficiently create a profile that gives a correspondence between colorimetric values and image data.

It is a block diagram which shows the structural example of the computer concerning embodiment of this invention. It is a figure which shows the color chart CC. It is a block diagram which shows the software structural example of the program run in the computer of FIG. It is a flowchart which shows the flow of a single lattice 3D-LUT creation process. It is a figure which shows the structural example of single grating | lattice 3D-LUT. It is a figure which shows the mode of normalization. It is a flowchart which shows the flow of a brightness function creation process. It is a figure explaining a brightness function creation process. It is another figure explaining a brightness function creation process. It is another figure explaining a brightness function creation process. It is another figure explaining a brightness function creation process. It is another figure explaining a brightness function creation process. It is a figure explaining HSB weighting coefficient creation processing. It is a block diagram which shows the structural example of the imaging | photography data reproduction part of FIG. FIG. 4 is a block diagram illustrating a configuration example of a colorimetric data reproduction unit in FIG. 3. It is a figure which shows the example of a display of UI screen.

  Embodiments of the present invention will be described below. Correspondences between the configuration requirements of the present invention and the embodiments described in the detailed description of the present invention are exemplified as follows. This description is to confirm that the embodiments supporting the present invention are described in the detailed description of the invention. Accordingly, although there are embodiments that are described in the detailed description of the invention but are not described here as embodiments corresponding to the constituent elements of the present invention, It does not mean that the embodiment does not correspond to the configuration requirements. Conversely, even if an embodiment is described here as corresponding to a configuration requirement, that means that the embodiment does not correspond to a configuration requirement other than the configuration requirement. It's not something to do.

An image processing apparatus according to one aspect of the present invention includes:
In an image processing apparatus for creating a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space,
Lookup that creates a lookup table that defines the correspondence between the input color space and the target color space based on the correspondence between the colorimetric values of multiple color patches and the input values obtained by inputting color patches with an image input device Table creation means (for example, parameter creation unit 72 and profile creation unit 73 in FIG. 3);
A lightness function creating means (for example, a lightness function creating unit 85 in FIG. 3) for creating a lightness function for converting the lightness of the input value defined in the lookup table so that the color difference from the colorimetric value is minimized; ,
Based on a lookup table to which the brightness function is applied, and a profile creation means for creating a profile (for example, the profile creation unit 73 in FIG. 3),
The brightness function creation means
A knee point in the brightness function is determined based on the average brightness function knee angle of the already created brightness function and the output value of the knee point (for example, FIGS. 8 to 12);
Lookup table creation means
A matrix that defines the correspondence between the input color space and the target color space is created based on the correspondence between the colorimetric value and the input value (for example, the matrix creation unit 83 in FIG. 3).
Create a single-grid three-dimensional lookup table storing the correspondence between the input color space defined by the matrix and the target color space (for example, the single-grid 3D-LUT creation unit 84 in FIG. 3),
The lightness of each color value of the target color space defined in the single-grid three-dimensional lookup table is normalized based on the white lightness obtained by converting the white point of the input color space using the matrix (for example, FIG. 3 Single lattice 3D-LUT creation unit 84),
A multi-grid lookup table is created by applying a single-grid three-dimensional lookup table (for example, multi-grid 3D-LUT creation unit 91, single-grid 3D-LUT application unit 92 in FIG. 3),
A lightness function is applied to the multi-grid lookup table (for example, the lightness function application unit 93 in FIG. 3).
It is characterized by that.

With respect to the weighting coefficient whose hue and saturation are corrected in accordance with the magnitude, a weighting coefficient creating means (for example, FIG. 3) creates a weighting coefficient having a magnitude inversely proportional to the magnitude of the output value in the nonlinear portion of the brightness function. HSB weighting factor generator 86)
Further comprising
The profile creation unit can create a profile based on the lookup table to which the weighting factor created by the weighting factor creation unit is applied (for example, the colorimetric color correction unit 94 in FIG. 3).

Display control means (for example, UI unit 76 in FIG. 3) for displaying an operation screen having an input unit (for example, buttons 211 to 217 in FIG. 16) that receives an output instruction of a matrix, a brightness function, or a weighting factor;
In accordance with an instruction received by the input unit, it may further include parameter output means (for example, parameter output unit 74 in FIG. 3) for outputting a matrix, a brightness function, and a weighting coefficient.

[Description of computer 1, DSC2, colorimeter 3]
FIG. 1 is a block diagram showing a configuration of a computer 1 according to an embodiment of the present invention.

  A digital still camera (DSC) 2 and a colorimeter 3 are connected to the computer 1 via a USB cable. The DSC 2 generates image data in which the color of each pixel constituting the photographed image is expressed in 256 gradations with respect to the three primary colors of RGB, and supplies the image data to the computer 1 via the USB cable. The colorimeter 3 includes a spectrophotometer, and supplies coordinate values (YXZ values) in the CIEExyz color system based on an output signal of the spectrophotometer to the computer 1 via a USB cable.

  In the case of this example, the DSC 2 captures a color chart CC composed of a plurality of color patches CP shown in FIG. 2, and the colorimeter 3 measures the same color chart CC. That is, an RGB value (hereinafter referred to as a camera RGB value) obtained as a result of imaging the color chart CC and an XYZ value (hereinafter referred to as a colorimetric XYZ value) as a colorimetric value of each color patch CP are respectively calculated by a computer. 1 is supplied.

  The computer 1 uses the camera RGB values from the DSC 2 and the colorimetric XYZ values from the colorimeter 3 to determine the camera RGB values defined in the input color space depending on the DSC 2 and the non-device-dependent target color space. A process of creating a profile that defines the correspondence relationship between the two (hereinafter referred to as profile creation process) is executed. The profile created here is used when the color indicated by each pixel of the image data is converted into a color value in a non-device-dependent color space. In particular, since DSC 2 is an image input device, this profile is called a source ICC profile.

[Description of Configuration Example of Computer 1]
The computer 1 includes a CPU 11, a RAM 12, a ROM 13, a hard disk drive (HDD) 14, a general-purpose interface (GIF) 15, a video interface (VIF) 16, an input interface (IIF) 17, and a bus 18. The bus 18 realizes data communication between the elements 11 to 17 constituting the computer 1, and communication is controlled by a chip set (not shown).

  The HDD 14 stores program data 14a for executing various programs including an operating system (OS), and the CPU 11 executes calculations according to the program data 14a while expanding the program data 14a in the RAM 12. . Further, the HDD 14 stores a conversion profile 14b that defines color conversion rules between non-device-dependent color spaces. According to the conversion profile 14b, the XYZ space, CIELAB space, and sRGB space can be mutually converted.

  The general-purpose interface (GIF) 15 provides an interface conforming to the USB standard, for example, and controls communication between the external DSC 2 and the colorimeter 3.

  A video interface (VIF) 16 connects the computer 1 to an external display 19 and displays an image on the display 19. An input interface (IIF) 17 connects the computer 1 to an external keyboard 20 and a mouse 21, and takes input signals from the keyboard 20 and the mouse 21 into the computer 1.

  FIG. 3 is a block diagram illustrating a configuration example of software of a program executed in the computer 1. In the figure, an image input device driver 51, a colorimeter driver 52, and a profile creation application (APL) 53 are executed.

  The image input device driver 51 controls the DSC 2, inputs camera RGB values output from the DSC 2, and stores them in the HDD 14 as RGB files. The camera RGB values are acquired by performing a decoding process (not shown) on the photographic data.

  The colorimeter driver 52 controls the colorimeter 3, inputs the colorimetric XYZ values output from the colorimeter 3, and stores them in the HDD 14 as an XYZ file.

  The image input device driver 51 and the colorimeter driver 52 are incorporated in, for example, an operating system (OS).

  The APL 53 executes profile creation processing. Here, the profile creation for DSC 2 will be described. However, the present invention is not limited to this, and the present invention can be applied to the case of creating a profile for other image input devices such as a scanner, a film scanner, and a digital video camera. .

[Detailed description of APL53]
The APL 53 includes a control unit 71, a parameter creation unit 72, a profile creation unit 73, and a parameter output unit 74 having a related data generation unit 75 and a UI unit 76.

[Detailed Description of Control Unit 71]
The control unit 71 controls each unit in accordance with an operation on a UI (User Interface) screen displayed by a UI unit 76 described later. The parameter creation unit 72 creates various color reproduction parameters according to the control of the control unit 71. The profile creation unit 73 creates a profile by appropriately using the color reproduction parameter created by the parameter creation unit 72 according to the control of the control unit 71. The parameter output unit 74 appropriately outputs the color reproduction parameter created by the parameter creation unit 72 according to the control of the control unit 71. The related data generation unit 75 reproduces the camera RGB values and the colorimetric XYZ values without using a profile.

[Description of Parameter Creation Unit 72]

The parameter creation unit 72 includes a data acquisition unit 81, a white balance coefficient calculation unit 82, a matrix creation unit 83, a single lattice 3D-LUT creation unit 84, a brightness function creation unit 85, and an HSB weight coefficient creation unit 86. Has been.
[Description of Data Acquisition Unit 81]
The data acquisition unit 81 acquires an RGB file and an XYZ file according to the control of the control unit 71. The data acquisition unit 81 also associates the acquired camera RGB values (input values) as RGB files with the colorimetric XYZ values (colorimetric values) as XYZ files as shown in FIG. Remember me.

[Description of White Balance Coefficient Calculation Unit 82]
The white balance coefficient calculation unit 82 adjusts the gain of the camera RGB values so that the gray axis where R = G = B corresponds to shooting under the CIE-D50 light source.

[Description of Matrix Creation Unit 83]
The matrix creation unit 83 calculates a matrix M that defines the correspondence between camera RGB values and XYZ values.

Specifically, the correspondence between camera RGB values and XYZ values defined by the matrix M is expressed by the following equation (1). In the formula (1), a11 to a33 represent matrix elements constituting the matrix M.

  The matrix M minimizes the square error between the correspondence between the camera RGB values and the colorimetric XYZ values indicated in the measurement data MD and the correspondence between the camera RGB values and the XYZ values defined by the matrix M. Calculated by the matrix least squares method.

[Description of Single Lattice 3D-LUT Creation Unit 84]
The single lattice 3D-LUT creation unit 84 creates a single lattice 3D lookup table (single lattice 3D-LUT). FIG. 4 is a flowchart showing the flow of the single lattice 3D-LUT creation process in the single lattice 3D-LUT creation unit 84.

  In step S11, the single grid 3D-LUT creation unit 84 creates a single grid 3D-LUT in which the correspondence between the camera RGB value and the colorimetric XYZ value is stored at each vertex.

  Specifically, the eight vertices of the cubic color space of the camera RGB values [K (Black), R (Red), G (Green), B (Blue), C (Cyan), M (Magenta), Y (Yellow)] , W (White)] are sequentially substituted into the above equation (1), and XYZ values corresponding to these are calculated. The 8 RGB vertices are associated with the camera RGB values and XYZ values at each vertex.

  FIG. 5 is a diagram schematically showing a single lattice 3D-LUT. Thus, the correspondence between the camera RGB value and the XYZ value is stored in each of the eight vertices of the single lattice 3D-LUT.

  In step S12, the single lattice 3D-LUT creation unit 84 adjusts the single lattice 3D-LUT so that the color difference is minimized.

  Specifically, the camera RGB values stored in the measurement data MD are converted into XYZ values with reference to the single grid 3D-LUT. The XYZ value converted here is appropriately referred to as a converted XYZ value.

  Next, the color difference ΔE between the converted XYZ value and the colorimetric XYZ value stored in the measurement data MD (the XYZ value obtained as a result of colorimetry by the colorimeter 3) is calculated. In this example, the color difference ΔE is calculated by the CIE_Δ1994 color difference formula. In the case of this example, the converted XYZ values and the colorimetric XYZ values are converted into L * a * b * values by the conversion profile 14b, and then these color differences ΔE are calculated. The color difference ΔE is calculated for each color patch CP.

  Then, the XYZ values associated with six vertices [R, G, B, C, M, Y] excluding Bk and W among the vertices of the single lattice 3D-LUT are adjusted so that the color difference ΔE is minimized. Is done. Specifically, the predetermined hue adjustment amount (ΔH value) and the saturation adjustment amount (ΔC value) are determined so that the color difference ΔE becomes small. Then, the ΔH value and ΔC value are converted into the adjustment amount of the XYZ value, and the XYZ values at the six vertices are adjusted.

  When the single grid 3D-LUT is adjusted so that the color difference is minimized in this way, in step S13, the single grid 3D-LUT creation unit 84 normalizes the single grid 3D-LUT based on the white lightness. To do.

  Specifically, the YW value (white brightness) associated with the white vertex [W] having the camera RGB value of 255 in the adjusted single lattice 3D-LUT is acquired. Then, each Y value associated with each grid point of the single grid 3D-LUT is uniformly divided by the white brightness YW. As a result, the Y value associated with each grid point is normalized to 0-1.

  In the XYZ space, the Y value is a value between 0 and 1, but since the matrix M is calculated without considering the dynamic range of the image sensor of DSC2, the YW value may exceed 1. Therefore, the Y value associated with each lattice point is normalized to 0 to 1 by uniformly dividing each Y value associated with each lattice point of the single lattice 3D-LUT by the whiteness YW. can do.

  FIG. 6 is a diagram schematically showing this normalization process. FIG. 6 shows the Y values before and after normalization associated with the grid points on the gray axis where the camera RGB values are R = G = B. At the grid points in the high brightness area, the Y value before normalization (dotted line in the figure) exceeds 1, but by normalization, the Y value (solid line in the figure) of each grid point falls within the range of 0 to 1. Will fit. By doing so, the Y value of the converted XYZ value converted by the single lattice 3D-LUT does not exceed 1, and as a result, overexposure can be prevented. Since the brightness is corrected downward as a whole, the converted XYZ value converted by the single grid 3D-LUT shows a darker color than the original.

  In this way, a single lattice 3D-LUT is created.

[Description of Lightness Function Creation Unit 85]
The brightness function creation unit 85 creates a brightness function. FIG. 7 is a flowchart showing the flow of the brightness function generation process.

  In step S <b> 21, the lightness function creating unit 85 adjusts the relationship of input value = output value in the relationship between the input value and the output value in lightness so that the color difference from the colorimetric data is minimized.

  Specifically, an output value that minimizes the color difference from the colorimetric data is obtained while using the camera RGB value of the single-grid 3D-LUT normalized based on white brightness as an input value. For example, camera RGB values stored in the measurement data MD are converted into XYZ values with reference to the single grid 3D-LUT. Each of the converted XYZ value and the colorimetric XYZ value is converted into an L * a * b * value by the conversion profile 14b, and the color difference ΔE is calculated. Then, the camera RGB values are adjusted so that the color difference ΔE becomes the minimum value (for example, 0 value).

  FIG. 8 is a diagram schematically showing this adjustment process. In the figure, the dotted line indicates the correspondence relationship between the input value and the output value, and the solid line indicates the relationship between the input value and the output value that minimizes the color difference from the colorimetric data. In the case of this example, by obtaining this gain coefficient, the relationship between the input value and the output value that minimizes the color difference from the colorimetric data can be obtained. In the following, the relationship between the input value and the output value that minimizes the color difference from the colorimetric data is referred to as a linear brightness function BF ′ as appropriate.

  In the processing from step S22 to step S24, the brightness function creation unit 85 detects a knee point.

  That is, in step S22, the brightness function creation unit 85 determines the angle of the knee point NPh in the average brightness function BFh of the plurality of brightness functions BF already created for the DSC2 model in the linear brightness function BF ′ obtained in step S21. A point (hereinafter referred to as a reference point SP1) having an angle coincident with (hereinafter referred to as a knee angle NPr) is detected.

  The left side of FIG. 9 shows the knee point NPh and the knee angle NPr in the average brightness function BFh. In this case, as shown on the right side of FIG. 9, the point on the straight line indicating the linear brightness function BF ′ and the angle of the straight line connecting the maximum input / output value point becomes the knee angle NPr. It is detected as a point SP1.

  Next, in step S23, the brightness function creation unit 85 corresponds to the output value corresponding to the output value of the knee point NPh of the average brightness function BFh in the linear brightness function BF ′ obtained in step S21 (hereinafter referred to as a reference point). (Referred to as SP2). The left side of FIG. 10 shows the output value NPho of the knee point NPh of the average brightness function BFh. In this case, as shown on the right side of FIG. 10, a point where the output value = NPho on the straight line indicating the linear brightness function BF ′ is detected as the reference point SP2. The knee angle NPr and output value of the knee point NPh in the average brightness function BFh are stored in the HDD 14, for example.

  When the reference point P1 and the reference point P2 are detected in this way, in step S24, the brightness function creating unit 85, as shown in FIG. 11, shows a straight line reference point SP1 and a reference point SP2 indicating the linear brightness function BF ′. The intermediate point is set as the knee point NP.

  When the knee point is detected in the processing from step S22 to step S24 as described above, in step S25, the brightness function creating unit 85, as shown in FIG. 12, from the origin, the knee point NP, and the origin to the knee point NP. The input value and output value and the maximum input / output value at points at predetermined intervals are acquired as the brightness function BF.

  In this way, the brightness function BF is created.

  Since the single-grid 3D-LUT is normalized based on the whiteness brightness, the brightness is corrected downward as a whole. As described above, the brightness function is adjusted so that the color difference from the colorimetric data is minimized. As a result, it is possible to approximate the original gradation reproduction.

  In addition, as described above, the knee point NP is determined using the already created brightness function, so that it is possible to eliminate variations among users.

[Description of HSB Weighting Factor Creating Unit 86]
The HSB weighting coefficient creating unit 86 performs correction to bring the hue and saturation in the non-linear part closer to the colorimetric value. Specifically, as shown in FIG. 13, the brightness function created by the brightness function creation unit 85 is interpolated by a spline curve passing through the point corresponding to the origin, knee point, and maximum input / output value.

  Then, the HSB weighting coefficient having a magnitude inversely proportional to the magnitude of the output value is assigned to the brightness coefficient (that is, the non-linear portion) from the knee point to the maximum input / output value. For example, the largest weighting factor (for example, 1) is assigned to the knee point NP, the smaller weighting factors are assigned to the subsequent points, and the maximum input / output value is the highest. A small weighting factor (eg, 0) is assigned.

[Description of Profile Creation Unit 73]
The profile creation unit 73 creates a lookup table and a profile based on the single lattice 3D-LUT, the brightness function, and the HSB weighting factor (that is, the color reproduction parameter) supplied from the parameter creation unit 72 according to the control of the control unit 71. To do. The profile creation unit 73 includes a multi-grid 3D-LUT creation unit 91, a single-grid 3D-LUT application unit 92, a brightness function application unit 93, a colorimetric color correction unit 94, and a mode color correction unit 95. .

  The multi-grid 3D-LUT creation unit 91 is a single-grid 3D-, which is a cubic color space of RGB values and whose eight vertices are associated with [K, R, G, B, C, M, Y, W]. A multi-lattice 3D-LUT is created by creating an LUT and equally dividing by multi-grid points.

  The single lattice 3D-LUT application unit 92 adds the single lattice 3D-LUT created by the single lattice 3D-LUT creation unit 84 of the parameter creation unit 72 to the multi lattice 3D-LUT created by the multi lattice 3D-LUT creation unit 91. Apply the LUT. Specifically, the RGB value and the XYZ value associated with each lattice point of the multi-grid 3D-LUT are calculated by performing tetrahedral interpolation on the single lattice 3D-LUT. Then, the calculated RGB value and XYZ value are associated with each other and stored in each lattice point of the multi-grid 3D-LUT.

  The single lattice 3D-LUT application unit 92 uses, as a profile, the multi-lattice 3D-LUT obtained as a result of applying the single lattice 3D-LUT created by the single lattice 3D-LUT creation unit 84 under the control of the control unit 71. (Hereinafter referred to as a formal profile) and stored in the HDD 14.

  The lightness function applying unit 93 uniformly applies the lightness function generated by the lightness function creating unit 85 to the RGB values associated with each lattice point of the formal profile. That is, the RGB value associated with each lattice point of the formal profile is adjusted by the brightness function. As a result, the multi-grid 3D-LUT once has a dark reproducibility by the normalization based on the white lightness described above, but by applying the lightness function generated by the lightness function creating unit 85, It is possible to approximate the gradation reproduction.

  The lightness function application unit 93 causes the HDD 14 to store a profile obtained by applying the lightness function to the formal profile (hereinafter referred to as a first colorimetry mode profile) according to the control of the control unit 71.

  The colorimetric color correction unit 94 corrects the first colorimetric mode profile based on the HSB weighting factor created by the HSB weighting factor creation unit 86 of the parameter creation unit 72. Specifically, the colorimetric value is converted into the HSB color space, the color-reproduced RAW data is converted into the HSB color space, and each difference in the HSB is obtained. Then, according to the HSB weighting factor, the HSB difference in the non-linear portion is corrected so as to be minimized.

  In this way, the weighting coefficient whose magnitude is inversely proportional to the output value in the non-linear part is corrected so that the hue and saturation of the non-linear part are close to the colorimetric value in the HSB color space, thereby suppressing the occurrence of pseudo contours. can do.

  The colorimetric color correction unit 94 corrects the profile obtained as a result of correcting the first colorimetry mode profile based on the HSB weighting coefficient under the control of the control unit 71 (hereinafter referred to as the second colorimetry mode profile). It is stored in the HDD 14.

  The mode color correction unit 95 creates a standard mode profile, a person mode profile, and a landscape mode profile by performing color correction for each mode on the second colorimetric mode profile.

  For example, a standard mode profile is created by applying an RGB contrast curve to the second colorimetric mode profile. A person mode profile is created by applying an L * curve to the second colorimetric mode profile and adjusting the skin color to be bright. In addition, a landscape mode profile is created by applying an RGB contrast curve to the second colorimetric mode profile and adjusting the blue and green colors to be vivid.

[Description of Parameter Output Unit 74]
The parameter output unit 74 is a RGB file or XYZ file acquired by the data acquisition unit 81 of the parameter creation unit 72 and a white balance coefficient calculated at the time of gain adjustment by the white balance coefficient calculation unit 82 according to the control of the control unit 71. The matrix M created by the matrix creation unit 83, the single grid 3D-LUT created by the single grid 3D-LUT creation unit 84, the brightness function created by the brightness function creation unit 85, or the HSB weight coefficient creation unit 86 The created HSB weight coefficient is stored in the HDD 14 as a parameter file.

[Description of Related Data Generation Unit 75]
The related data generation unit 75 includes a photographic data reproduction unit 101 and a colorimetric data reproduction unit 102. The imaging data reproduction unit 101 reproduces the imaging data of DSC 2 based on the brightness function and the HSB weight coefficient without using a lookup table. The colorimetric data reproduction unit 102 reproduces the colorimetric data of the colorimeter 3 based on the brightness function and the HSB weight coefficient without using a lookup table.

[Description of Imaging Data Reproducing Unit 101]
FIG. 14 is a block diagram illustrating a configuration example of the shooting data reproduction unit 101. The shooting data reproduction unit 101 includes a hue / saturation compensation unit 111, a colorimetric color correction unit 112, and a mode color correction unit 113.

  The hue / saturation compensation unit 111 performs brightness conversion on the camera RGB value as an RGB file while compensating for hue and saturation based on the brightness function created by the brightness function creation unit 85.

  The colorimetric color correction unit 112 applies an HSB weighting factor to the camera RGB values whose brightness has been converted by the hue saturation compensation unit 111.

  The mode color correction unit 113 operates in the same manner as the mode color correction unit 95 of the profile creation unit 73, and for each camera RGB value processed by the hue / saturation compensation unit 111 or the colorimetric color correction unit 112, for each mode. Perform color correction.

  As described above, since the camera RGB values can be reproduced without using the profile and the image for each mode can be referred to, the image is directly reproduced from the camera RGB values and created by the profile creation unit 73. It is possible to confirm a difference from an image whose color is reproduced by a profile.

[Description of Colorimetric Data Reproducing Unit 102]
The colorimetric data reproduction unit 102 includes a mode color correction unit 121 as shown in FIG. The mode color correcting unit 121 operates in the same manner as the mode color correcting unit 95 of the profile creating unit 73, and for the colorimetric XYZ value as an XYZ file or the colorimetric XYZ value whose brightness has been adjusted by the brightness function. Apply color correction for each mode.

  As described above, the colorimetric XYZ values can be reproduced without using a profile, and the image for each mode can be referred to. Therefore, an image directly reproduced from the colorimetric XYZ values and the profile creating unit 73 It is possible to confirm the difference from the color reproduced image by the profile obtained.

[Description of UI Unit 76]
FIG. 16 is a diagram illustrating an example of the UI screen 201 displayed by the UI unit 76. When the user operates various buttons on the UI screen 201, acquisition of color reproduction parameters and creation of various profiles can be instructed.

  A button 211 is operated when designating an RGB file or an XYZ file to be captured. Buttons 212 to 217 described later can be operated after the button 211 is operated and the RGB file and the XYZ file are captured.

  The button 212 is operated when acquiring the white balance coefficient. The button 213 is operated when acquiring the matrix M. The button 214 is operated when acquiring the single lattice 3D-LUT. The button 215 is operated when acquiring the brightness function. The button 216 is operated when acquiring the HSB weight coefficient. The button 217 is operated when creating a profile for each mode.

  The operation of the APL 53 according to the operation of each button will be described. When the button 211 is operated to specify the RGB file and the XYZ file, the control unit 71 controls the data acquisition unit 81 of the parameter creation unit 72 to acquire the specified RGB file and XYZ file.

  When the button 212 is operated, the control unit 71 controls the white balance coefficient calculation unit 82 to calculate the white balance coefficient from the camera RGB value as the RGB file acquired by the data acquisition unit 81. The control unit 71 controls the parameter output unit 74 to store the calculated white balance coefficient in the HDD 14.

  When the button 213 is operated, the control unit 71 controls the white balance coefficient calculation unit 82 to calculate the white balance coefficient, and controls the matrix creation unit 83 to create the matrix M. The control unit 71 controls the parameter output unit 74 to store the created matrix M in the HDD 14.

  When the button 214 is operated, the control unit 71 controls the white balance coefficient calculation unit 82 to calculate the white balance coefficient, controls the matrix creation unit 83 to create the matrix M, and creates the single lattice 3D-LUT. The unit 84 is controlled to create a single lattice 3D-LUT. The control unit 71 controls the parameter output unit 74 to store the created single lattice 3D-LUT in the HDD 14.

  When the button 215 is operated, the control unit 71 controls the white balance coefficient calculation unit 82 to calculate the white balance coefficient, controls the matrix creation unit 83 to create the matrix M, and the single lattice 3D-LUT creation unit. 84 is controlled to create a single lattice 3D-LUT, and the brightness function creation unit 85 is controlled to create a brightness function. The control unit 71 controls the parameter output unit 74 to store the created brightness function in the HDD 14.

  When the button 216 is operated, the control unit 71 controls the white balance coefficient calculation unit 82 to calculate the white balance coefficient, controls the matrix creation unit 83 to create the matrix M, and the single lattice 3D-LUT creation unit. 84 is controlled to create a single-grid 3D-LUT, the brightness function creation unit 85 is controlled to create a brightness function, and the HSB weight coefficient creation unit 86 is controlled to create an HSB weight coefficient. The control unit 71 controls the parameter output unit 74 to store the created HSB weight function in the HDD 14.

  When the button 217 is operated, the control unit 71 controls the parameter creation unit 72 to operate each unit and also controls the profile creation unit 73 to create color reproduction parameters created in each unit of the parameter creation unit 72. Create each profile based on

  As described above, since various color reproduction parameters can be acquired, the color reproduction parameters can be used in other application programs. Further, by simply operating the button 217, color reproduction parameters are automatically calculated and various profiles are created, so that profiles can be created quickly.

[Other embodiments]
The program describing the processing contents of the functions described above can be recorded on a computer-readable recording medium. Examples of the computer-readable recording medium include a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Examples of the magnetic recording device include a hard disk device (HDD), a flexible disk (FD), and a magnetic tape. Optical discs include DVD (Digital Versatile Disc), DVD-RAM, CD-ROM (Compact Disc ROM), CD-R (Recordable) / RW (ReWritable), and the like. Magneto-optical recording media include MO (Magneto-Optical disc).

  When distributing the program, for example, portable recording media such as a DVD and a CD-ROM in which the program is recorded are sold. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read a program directly from a portable recording medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

  In the above embodiment, a part of the configuration realized by hardware may be replaced with software, and conversely, a part of the configuration realized by software may be replaced by hardware. .

  The process shown in the flowchart is executed in parallel or individually even if the steps are not necessarily processed in time series, as well as processes performed in time series in the order described. It includes processing that. Further, the processing can be omitted as appropriate.

  51 image input device driver, 52 colorimeter driver, 53 APL, 71 control unit, 72 parameter creation unit, 73 profile creation unit, 74 parameter output unit, 75 related data generation unit, 76 UI unit, 81 data acquisition unit, 82 White balance coefficient calculation unit, 83 matrix creation unit, 84 single grid 3D-LUT creation unit, 85 brightness function creation unit, 86 HSB weight coefficient creation unit, 91 multi-grid 3D-LUT creation unit, 92 single grid 3D-LUT application unit , 93 brightness function application unit, 94 colorimetric color correction unit, 95 mode color correction unit, 101 photographing data reproduction unit, 102 colorimetric data reproduction unit, 111 hue saturation compensation unit, 112 colorimetric color correction unit, 113 mode color correction unit 121 mode color correction unit, 201 UI screen, 211-217 buttons

Claims (6)

In an image processing apparatus for creating a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space,
A lookup table for defining a correspondence relationship between the input color space and the target color space based on a correspondence relationship between colorimetric values of a plurality of color patches and an input value obtained by inputting the color patch by the image input device; A lookup table creation means to create;
A lightness function creating means for creating a lightness function for converting the lightness of the input value defined in the lookup table so that the color difference from the colorimetric value is minimized;
Profile creation means for creating the profile based on the lookup table to which the brightness function is applied;
The brightness function creating means is:
The knee point in the brightness function is determined based on the knee angle of the average brightness function of the brightness function already created and the output value of the knee point,
The lookup table creation means
Creating a matrix that defines the correspondence between the input color space and the target color space based on the correspondence between the colorimetric value and the input value;
Creating a single-grid three-dimensional lookup table storing a correspondence relationship between the input color space defined by the matrix and the target color space;
Normalizing the lightness of each color value of the target color space defined in the single-grid three-dimensional lookup table based on the white lightness obtained by converting the white point of the input color space using the matrix;
A multi-grid lookup table is created by applying the single-grid three-dimensional lookup table,
An image processing apparatus, wherein the brightness function is applied to the multi-grid lookup table. The image processing apparatus according to claim 1.
A weighting factor creating means for creating a weighting factor having a magnitude inversely proportional to the magnitude of the output value in the nonlinear portion of the brightness function for the weighting factor whose hue and saturation are corrected according to the magnitude. ,
The image processing apparatus, wherein the profile creation unit creates the profile based on the lookup table to which the weighting factor created by the weighting factor creation unit is applied. The image processing apparatus according to claim 2,
Display control means for displaying an operation screen having an input unit for receiving an instruction to output the matrix, the brightness function, or the weighting factor;
An image processing apparatus, further comprising: parameter output means for outputting the matrix, the brightness function, and the weighting coefficient in accordance with an instruction received by the input unit. The image processing apparatus according to claim 1.
The look-up table creation means does not apply the brightness function. In an image processing method of an image processing apparatus for creating a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space,
A lookup table for defining a correspondence relationship between the input color space and the target color space based on a correspondence relationship between colorimetric values of a plurality of color patches and an input value obtained by inputting the color patch by the image input device; A lookup table creation step to create,
A brightness function creating step for creating a brightness function for converting the brightness of the input value defined in the lookup table so that the color difference from the colorimetric value is minimized;
Creating a profile based on the look-up table to which the lightness function is applied; and
The brightness function creation step
The knee point in the brightness function is determined based on the knee angle of the average brightness function of the brightness function already created and the output value of the knee point,
The above lookup table creation step
Creating a matrix that defines the correspondence between the input color space and the target color space based on the correspondence between the colorimetric value and the input value;
Creating a single-grid three-dimensional lookup table storing a correspondence relationship between the input color space defined by the matrix and the target color space;
Normalizing the lightness of each color value of the target color space defined in the single-grid three-dimensional lookup table based on the white lightness obtained by converting the white point of the input color space using the matrix;
A multi-grid lookup table is created by applying the single-grid three-dimensional lookup table,
An image processing method, wherein the brightness function is applied to the multi-grid lookup table. A program for causing a computer to execute image processing for creating a profile that defines a correspondence relationship between an input color space used by an image input device and a predetermined target color space,
A lookup table for defining a correspondence relationship between the input color space and the target color space based on a correspondence relationship between colorimetric values of a plurality of color patches and an input value obtained by inputting the color patch by the image input device; A lookup table creation step to create,
A brightness function creating step for creating a brightness function for converting the brightness of the input value defined in the lookup table so that the color difference from the colorimetric value is minimized;
Creating a profile based on the look-up table to which the lightness function is applied; and
The brightness function creation step
The knee point in the brightness function is determined based on the knee angle of the average brightness function of the brightness function already created and the output value of the knee point,
The above lookup table creation step
Creating a matrix that defines the correspondence between the input color space and the target color space based on the correspondence between the colorimetric value and the input value;
Creating a single-grid three-dimensional lookup table storing a correspondence relationship between the input color space defined by the matrix and the target color space;
Normalizing the lightness of each color value of the target color space defined in the single-grid three-dimensional lookup table based on the white lightness obtained by converting the white point of the input color space using the matrix;
A multi-grid lookup table is created by applying the single-grid three-dimensional lookup table,
A program for causing a computer to execute image processing, wherein the brightness function is applied to the multi-grid lookup table.

Images