JP2011087018A - Color processor, and method therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To generate an image under an arbitrary observation light source of an article represented as a color-adjusted image. <P>SOLUTION: The image and the observation light source are set (S11), the color of the image is adjusted (S13-S15), the spectral data of the article represented as the color-adjusted image are calculated (S17, S18), and the image of the article is generated on the basis of the calculated spectral data and the set observation light source (S19, S20). <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to color processing for estimating the perceived color of an article under an observation light source.

  Product design is performed using computer graphics (CG) software running on computer equipment (PC). Then, a material or paint corresponding to the color of the designed product is selected. Considering that the product is not observed with a specific light source, a spectroscopic color design is required. However, it is difficult to perform color design based on the spectral distribution. Generally, color design combining sRGB or CIELAB color spaces and specific light sources is performed.

  When materials or paints corresponding to RGB values or Lab values designed under a specific light source are selected, if the light source is different, it is perceived as a color different from the assumed color. In order to avoid this, it is possible to design the color in consideration of perceived colors under multiple light sources, but it is difficult to select materials and paints corresponding to RGB values and Lab values under multiple light sources. is there.

  Patent Document 1 discloses a technique for once converting a spectral reflectance image into a CIELAB space, retouching it in the CIELAB space, and inversely transforming it into a spectral reflectance. However, the inverse conversion to the spectral reflectance is a conversion from three dimensions (XYZ values) to one dimension (spectral values), and the solution is not uniquely determined. Therefore, Patent Document 1 outputs a value close to the spectral reflectance of the original image before retouching as a solution for inverse transformation.

  However, the technique disclosed in Patent Document 1 cannot guarantee the perceived color under other light sources. In addition, since a spectral reflectance image is required as an input image, even if it can be used to adjust the color of an existing product, it cannot be applied when designing a product from the beginning.

JP 2004-005566 A

  An object of the present invention is to generate an image under an arbitrary observation light source of an article represented by a color-adjusted image.

  The present invention has the following configuration as one means for achieving the above object.

  The color processing according to the present invention sets an image and an observation light source, adjusts the color of the image, calculates spectral data of an article represented by the color-adjusted image, and sets the calculated spectral data and the setting An image of the article is generated based on an observation light source.

  According to the present invention, it is possible to generate an image under an arbitrary observation light source of the article represented by the color-adjusted image.

1 is a block diagram illustrating a configuration example of a color processing apparatus according to Embodiment 1. FIG. 6 is a flowchart illustrating an example of color processing executed by the color processing apparatus. The figure which shows an example of UI displayed on a UI part. The figure which shows the example of a display of a spectral reflectance. The figure which shows the other example of a display of a spectral reflectance. The flowchart explaining the process of a spectrum estimation part. The flowchart explaining the process of a color value recalculation part. FIG. 3 is a block diagram illustrating a configuration example of a color processing apparatus according to a second embodiment. The flowchart explaining the process of a spectrum estimation part. The figure which shows an example of the table which a correspondence table holding part hold | maintains.

  Hereinafter, color processing according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Device configuration]
A configuration example of the color processing apparatus 11 according to the first embodiment will be described with reference to the block diagram of FIG.

  The UI unit 101 is a user interface (UI) display unit that allows the user to specify (or select) an image file or an observation light source and adjust the color of the image. The image input unit 102 is an interface for inputting an image file designated by the user. The image input unit 102 is, for example, a serial bus interface or a network interface, and inputs an image file designated by the user from an image input device, an image storage device, a storage medium, or the like connected to the serial bus or the network. Alternatively, the image input unit 102 is a card reader / writer, and inputs an image file designated by the user from a memory card inserted into the card reader / writer.

  The light source information storage unit 103 stores light source information of various observation light sources, and outputs light source information corresponding to the observation light sources specified by the user. The color adjustment unit 105 acquires an adjustment value set by the user. The spectral estimation unit 106 estimates the spectral reflectance corresponding to the adjustment value. The display unit 104 displays the spectral reflectance estimated by the spectral estimation unit 106. The color value recalculator 107 recalculates the color value from the spectral reflectance estimated by the spectral estimator 106.

  The matrix holding unit 108 is a non-volatile memory that holds a matrix for the spectral estimation unit 106 to estimate the spectral reflectance from the color value. The buffer memory 109 is a memory such as a RAM that temporarily stores calculation results during processing. The above-described components are connected to each other via a system bus 110.

[Color processing]
An example of color processing executed by the color processing apparatus 11 will be described with reference to the flowchart of FIG.

  The color processing apparatus 11 displays on the UI unit 101 a UI for the user to specify an image file and an observation light source and to perform color adjustment of the image (S11). When an image file and a light source are instructed by the UI unit 101 and the “image display” button 1003 is pressed (S12), the color processing apparatus 11 inputs the image file by the image input unit 102 (S13). Then, the light source information of the designated light source is read from the light source information storage unit 103 (S14), and an image obtained by performing color conversion corresponding to the light source on the original image of the image file is displayed on the UI (S15).

  An example of the UI displayed on the UI unit 101 will be described with reference to FIG. 3, and an example in which three types of high color rendering 3000K, three wavelengths 5000K, and ordinary 6500K, which are typical fluorescent light sources, are selected as light sources is shown. Yes.

  The image file name input unit 1001 is a field for inputting (or selecting) a file name of an original image before the user performs color adjustment, for example, a design image (or photographed image) of a product (article). The observation light source selection unit 1002 includes a check box for selecting an observation light source of the color adjustment target article, and can select a plurality of light sources. An “image display” button 1003 is a button for displaying the image indicated by the image file input by the image input unit 102 by the number of selected observation light sources.

  When an “image display” button 1003 is pressed, the color processing apparatus 11 displays on the image display unit 1004 an image obtained by performing color conversion corresponding to the light source on the original image for the number of selected observation light sources. The user can arbitrarily adjust each RGB value of the original image, for example, between −50% and + 50% by operating each slider (image adjustment unit 1005) of the RGB value provided in each of the image display units 1004. . Setting the R value to + 10% means changing a pixel value of a certain pixel of the original image, for example, RGB = (80, 50, 40) to (88, 50, 40). If the value after change is less than 0 or greater than 255, it is limited to 0 or 255.

  A “spectral estimation” button 1006 is a button for instructing estimation of spectral reflectance of an image after color adjustment. When the “spectral estimation” button 1006 is pressed, the color value corresponding to each observation light source is recalculated based on the estimated spectral reflectance, and the image of the article after color adjustment is displayed in the redisplay unit 1007, as will be described in detail later. Is displayed.

  The user compares the image before color adjustment displayed on the image display unit 1004 with the image after color adjustment displayed on the redisplay unit 1007, and compares the images after color adjustment displayed on the redisplay unit 1007. By comparison, it is determined whether or not the color is as expected or appropriate. If the color of the image of the article after color adjustment is not as expected or appropriate, the user presses the “Readjust” button 1008 to instruct the color readjustment. In addition, when the color of the image of the article after color adjustment is as expected and appropriate, the user presses an “output” button 1009 to instruct the output of the estimated spectral reflectance.

  When the “spectral estimation” button 1006 is pressed (S16), the color processing apparatus 11 adjusts the RGB value of the original image based on the adjustment value for each observation light source by the color adjustment unit 105 (S17). Then, as will be described in detail later, the spectral estimation unit 106 estimates the spectral reflectance of the article indicated by the adjusted image (S18). Then, the color value recalculation unit 107 calculates the color value under each observation light source based on the estimated spectral reflectance and the light source information of each observation light source, as will be described in detail later (S19). Then, based on the recalculated color value, the image after color adjustment is displayed on the redisplay unit 1007 (S20).

  Next, the color processing apparatus 11 repeats the determination whether the “readjustment” button 1008 has been pressed (S21) and the determination whether the “output” button 1009 has been pressed (S22). When the “readjustment” button 1008 is pressed, the process returns to step S17, and when the “output” button 1009 is pressed, the estimated spectral reflectance is displayed on the display unit 104 (S23).

  For the spectral reflectance display, the average value of the spectral reflectance of a predetermined region of the image may be displayed in a graph as shown in FIG. 4, for example, or the reflectance for each wavelength as shown in FIG. May be listed as The display form may be a monitor display or printing by a printer, or may be stored as a data file on a recording medium. Further, the display area of the spectral reflectance may be displayed on the re-display unit 1007 so that the display area can be moved by a pointing device not shown in the drawing. In this case, the color processing apparatus 11 updates the display on the display unit 104 as the display target area moves.

[Spectral Estimator]
The processing of the spectral estimation unit 106 will be described with reference to the flowchart of FIG.

The spectral estimation unit 106 acquires the adjusted image and the light source information of the observation light source stored in the buffer memory 109 (S81), and converts the RGB value of the acquired image into a CIEXYZ value (S82). If the RGB value is a color value in the sRGB color space defined by IEC61966-2-1, it is converted into an XYZ value using the following conversion matrix.
┌ ┐ ┌ ┐┌ ┐
│X│ │0.4124 0.3576 0.1805││R│
│Y│ = │0.2126 0.7152 0.0722││G│… (1)
│Z│ │0.0193 0.1192 0.9505││B│
└ ┘ └ ┘└ ┘

Next, the spectral estimation unit 106 acquires a conversion matrix G corresponding to the designated combination of observation light sources from the matrix holding unit 108 (S83). The conversion matrix G is a matrix for estimating the spectral reflectance from the XYZ values under a plurality of light sources. As in the UI shown in FIG. 3 as an example, a plurality of observation light sources can be selected from a plurality of types of light sources. The matrix holding unit 108 holds a conversion matrix G corresponding to each arbitrary combination of a plurality of types of light sources. For example, when high color rendering 3000K, three wavelengths 5000K, and normal 6500K are designated as the observation light source, a conversion matrix G corresponding to the combination of the three types of light sources is acquired. The conversion matrix G in the present embodiment is shown in the following equation.
┌ ┐
│g _1,1 … g _1,9 │
│g _2,1 … g _2,9 │
G = │::: │… (2)
│g _35,1 ... g _35,9 │
│g _36,1 … g _36,9 │
└ ┘

  The creation method of the conversion matrix G will be described later. As shown in an example in the equation (2), the conversion matrix G corresponding to the three types of light sources can display the XYZ values under the three types of light sources in the visible range (380 to 730 nm). This is a 36 × 9 matrix that converts to a spectral reflectance of 10 nm pitch. When there are two types of light sources, the matrix is 36 × 6, and when there are five types, the matrix is 36 × 15.

  As shown in the following equation, the spectral estimation unit 106 converts the XYZ values into spectral reflectances using the acquired conversion matrix G (S84), and stores the spectral reflectances in the buffer memory 109 (S85).

If all the pixels of the image after color conversion are converted from RGB values to XYZ values, and from XYZ values to spectral reflectances, an image represented by spectral reflectances can be obtained. However, when the size of the original image is large, calculating the spectral reflectance of all pixels increases the calculation load and the amount of memory used. In that case, create a reduced image of the original image, display the reduced image on the image display unit 1004, calculate the spectral reflectance for each pixel of the reduced image, and display the image after color adjustment of the redisplay unit 1007 By doing so, it is possible to reduce the computation load and the memory usage.
┌ ┐ ┌ ┐
│r ₃₈₀ │ │Xbb│
│r ₃₉₀ │ │Ybb│
│r ₄₀₀ │ │Zbb│
│ │ │ │Xtb│
│r ₅₅₀ │ = G│Ytb│… (3)
│ ： │ │Ztb│
│r ₇₁₀ │ │Xn│
│r ₇₂₀ │ │Yn│
│r ₇₃₀ │ │Zn│
└ ┘ └ ┘
Where (Xbb, Ybb, Zbb) is the XYZ value under the high color rendering fluorescent light source,
(Xtb, Ytb, Ztb) is the XYZ value under the three-wavelength fluorescent light source,
(Xn, Yn, Zn) is the XYZ value under a normal fluorescent light source.

Method for Creating Conversion Matrix Next, a method for creating the conversion matrix G for estimating the spectral reflectance from the CIXYZ values will be described.

First, the spectral reflectance R (λ) for a plurality of materials (paints and dyes) is measured or acquired. Then, when creating the conversion matrix G corresponding to the combination of the above-mentioned observation light sources, the XYZ values for the light sources of the high color rendering 3000K, the three wavelengths 5000K, and the normal 6500K are calculated by the following equations.
X = k∫R (λ) S (λ) x (λ) dλ
Y = k∫R (λ) S (λ) y (λ) dλ (4)
Z = k∫R (λ) S (λ) z (λ) dλ
Where S (λ) is the spectral radiance of the light source,
x (λ) y (λ) z (λ) is the CIE color matching function,
The integration range is 380-730mm.

The relationship among the spectral reflectances R (λ) of a plurality of materials, the XYZ values under a plurality of light sources, and the conversion matrix G can be expressed by the following equation when the number of materials is n.
o = Gv (5)
here,
┌ ┐
│o _380,1 ... g _{380, n} │
│g _390,1 … g _{390, n} │
o = │::: │
│g _720,1 … g _{720, n} │
│g _730,1 … g _{730, n} │
└ ┘
┌ ┐
│X _{bb, 1} … X _{bb, n} │
│Y _{bb, 1} … Y _{bb, n} │
│z _{bb, 1} … z _{bb, n} │
│X _{tb, 1} … X _{tb, n} │
v = │Y _{tb, 1} … Y _{tb, n} │
│z _{tb, 1} … z _{tb, n} │
│X _{n, 1} … X _{n, n} │
│Y _{n, 1} … Y _{n, n} │
│Z _{n, 1} … Z _{n, n} │
└ ┘

Therefore, the conversion matrix G can be obtained by the following equation using, for example, the least square method. As a method for obtaining the matrix G, for example, a Newton method or a DLS (dumped least square) method may be used.
G = ov ^T (vv ^T ) ^-1 … (6)
Here, T represents a transposed matrix.

  The conversion matrix G corresponding to any combination of the light sources created in this way is stored in the matrix holding unit 108.

[Color value recalculation section]
The processing of the color value recalculation unit 107 will be described with reference to the flowchart of FIG.

  The color value recalculation unit 107 acquires the light source information of the observation light source from the buffer memory 109 (S91), and acquires the estimated spectral reflectance (S92). Then, the XYZ values of each pixel are calculated from the spectral reflectance based on the light source information, and an XYZ image under the designated light source is generated (S93). For the calculation of the XYZ values, the above-described equation (4) is used.

Next, the color value recalculation unit 107 converts the XYZ image into an RGB image using the following equation (S94), and stores the RGB image corresponding to the designated light source in the buffer memory 109 (S95).
┌ ┐ ┌ ┐┌ ┐
│R│ │ 3.2406 -1.5372 -0.4986││X│
│G│ = │-0.9689 1.8758 0.0415││Y│… (7)
│B│ │ 0.0557 -0.2040 1.0570││Z│
└ ┘ └ ┘└ ┘

  Thus, the spectral reflectance of the article corresponding to the image after color adjustment is estimated, and the image before color adjustment and the image after color adjustment under a designated light source are displayed side by side. The user can easily determine whether the perceived color of the article after color adjustment under a designated light source is as expected or appropriate by referring to both images.

  Further, images of articles before and after color adjustment under a plurality of observation light sources are displayed side by side. The user can grasp how the color of the article before and after color adjustment is perceived under a plurality of observation light sources, and easily determine whether the color of the article is as expected or appropriate. can do.

  Further, since the spectral reflectance after color adjustment is displayed, the user can appropriately select a material suitable for the spectral reflectance. In other words, it is useful for selecting materials for colors designed by CG.

  The color processing according to the second embodiment of the present invention will be described below. Note that the same reference numerals in the second embodiment denote the same parts as in the first embodiment, and a detailed description thereof will be omitted.

  A configuration example of the color processing apparatus 11 according to the second embodiment will be described with reference to the block diagram of FIG. The color processing apparatus 11 according to the second embodiment includes a correspondence table holding unit 111 instead of the matrix holding unit 108 of the color processing apparatus 11 according to the first embodiment. The correspondence table holding unit 111 holds spectral reflectances corresponding to RGB values under a predetermined observation light source as a table. The color processing apparatus 11 of the second embodiment is different from the operation of the first embodiment in that the spectral reflectance estimation (S18) of the article indicated by the image after color adjustment by the spectral estimation unit 106 is different.

  The processing of the spectral estimation unit 106 will be described with reference to the flowchart of FIG. The spectral estimation unit 106 acquires the adjusted image and the light source information of the observation light source stored in the buffer memory 109 (S281), and converts the RGB value of the acquired image into an XYZ value (S282). If the RGB value is a color value in the sRGB color space defined by IEC61966-2-1, it is converted into an XYZ value using the conversion matrix of Equation (1).

  Next, the spectral estimation unit 106 refers to the spectral reflectance corresponding to the XYZ value under the designated observation light source from the table held by the correspondence table holding unit 111 (S283), and compares the spectral reflectances. Then, spectral reflectances whose shapes are close to each other are selected for each observation light source (S284).

  An example of a table held by the correspondence table holding unit 111 will be described with reference to FIG. In the table, spectral reflectances of a plurality of materials corresponding to XYZ values under a predetermined observation light source are recorded. A table can be created by converting an arbitrary spectral reflectance into an XYZ value according to the equation (4). For example, it is assumed that the spectral reflectances of 10 types of materials corresponding to the XYZ values under each observation light source are recorded in the table. Spectral estimator 106 selects the spectral reflectance characteristics that are close to each other for each observation light source by referring to the shape of the curve (spectral reflectance characteristics of the material) drawn by the spectral reflectance relative to the XYZ values for the specified observation light source. To do. Note that the shape may be determined by, for example, a spectral Euclidean distance.

  Next, the spectral estimation unit 106 averages the spectral reflectance characteristics selected for each observation light source (S285). For example, an average value is calculated for the spectral reflectance for each wavelength. Then, the averaged spectral reflectance is stored in the buffer memory 109 (S286).

  If conversion of RGB values to XYZ values and estimation of spectral reflectance based on XYZ values are performed for all pixels of the image after color conversion, an image represented by spectral reflectance can be obtained. However, as in the first embodiment, when the size of the original image is large, calculating the spectral reflectance of all the pixels increases the calculation load and the memory usage. In that case, create a reduced image of the original image, display the reduced image on the image display unit 1004, calculate the spectral reflectance for each pixel of the reduced image, and display the image after color adjustment of the redisplay unit 1007 By doing so, it is possible to reduce the computation load and the memory usage.

  In this way, by maintaining the correspondence relationship between the XYZ values under a predetermined observation light source and the spectral reflectance characteristics of a plurality of materials in advance, the spectral corresponding to the XYZ value can be obtained from the XYZ values under the predetermined observation light source. By estimating the reflectance, the spectral reflectance of the article indicated by the color-adjusted image can be estimated.

[Modification]
In the above embodiment, sRGB image data has been described as an example. However, the image data is not limited to sRGB, and may be data in other RGB spaces such as AdobeRGB. Furthermore, data of color appearance space such as CIELAB, CIECAM97, and CIECAM02 may be used instead of the RGB space.

  In the above-described embodiment, the example in which the spectral reflectance is calculated from the CIE tristimulus values (XYZ values) has been described. However, for example, RGB values may be directly converted into spectral reflectance. Further, color appearance spaces such as CIELAB, CIECAM97, and CIECAM02 may be used. In that case, the conversion matrix in the first embodiment and the table in the second embodiment may be used or created for the color space to be used.

  In the above-described embodiment, an example in which the user performs color adjustment on the original image under the observation light source has been described. However, for example, an image under an observation light source that has been color-adjusted in advance may be input. Also, it is possible to input only RGB values instead of images.

  Further, in the first embodiment, the example in which the conversion matrix G for converting the XYZ values into the spectral reflectance is held in the matrix holding unit 108 has been described. However, for example, the spectral estimation unit 106 may dynamically generate the conversion matrix G according to the observation light source designated by the user. In this case, the spectral radiance of each observation light source and a plurality of spectral reflectances which are learning data for creating the conversion matrix G are separately held.

  In the above embodiment, the spectral reflectance is described as 36-dimensional data obtained by sampling the visible range of 380 to 730 nm in units of 10 nm. However, for example, the wavelength range may be set to 380 to 780 nm. Also, the sampling interval may be 5 nm or 1 nm instead of 10 nm.

  In the above-described embodiment, an example in which the spectral reflectance is estimated has been described. However, the estimation target is not limited to spectral data such as spectral reflectance, but may be spectral data such as spectral transmittance and spectral radiance, for example.

  In the above-described embodiment, nine types of observation light sources have been described as an example. However, these are representative examples of light sources, and the observation light sources are not limited to the nine types. For example, the color temperature may be further subdivided so that it can be selected, or even light sources of the same waveform type may be further subdivided, for example, D50 and F8 fluorescent lamps.

  In the second embodiment, the example in which the XYZ values are converted into the spectral reflectance using the table has been described. It is not necessary to record the spectral reflectance corresponding to all the XYZ values in this table, the spectral reflectance corresponding to the XYZ values sampled at a predetermined interval is recorded, and the spectral reflectance corresponding to the unrecorded XYZ values. May be obtained by interpolation calculation.

  Further, in the above-described embodiment, an example in which the color value is recalculated from the estimated spectral reflectance and an image is displayed on the redisplay unit 1007 has been described. However, for example, the recalculated color value may be displayed as a numerical value as it is, or the difference between the original image color value and the recalculated color value may be displayed.

[Other Examples]
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, etc.) of the system or apparatus reads the program. It is a process to be executed.

Claims (8)

Setting means for setting an image and an observation light source;
Adjusting means for adjusting the color of the image;
Calculating means for calculating spectroscopic data of an article represented by the color-adjusted image;
A color processing apparatus comprising: a generating unit configured to generate an image of the article based on the calculated spectral data and the set observation light source.   2. The color processing apparatus according to claim 1, wherein the generation unit generates an image of the article by calculating a perceived color of the article under the observation light source from the spectral data.   2. The color processing apparatus according to claim 1, further comprising display means for displaying spectral data of a predetermined region of the image of the article.   The setting means can set a plurality of observation light sources, the adjustment means can adjust the color for each observation light source, and the calculation means uses the conversion matrix corresponding to a combination of the plurality of observation light sources. 2. The color processing apparatus according to claim 1, wherein spectral data is calculated.   The setting means can set a plurality of observation light sources, the adjustment means can adjust the color for each observation light source, and the calculation means can perform spectral data corresponding to a color value under each of the plurality of observation light sources. The color processing apparatus according to claim 1, wherein the spectral data is calculated using a table indicating A setting step for setting an image and an observation light source;
An adjustment step for adjusting the color of the image;
A calculation step of calculating spectral data of the article represented by the color-adjusted image;
A color processing method comprising: a generation step of generating an image of the article based on the calculated spectral data and the set observation light source.   6. A non-transitory computer-readable storage medium storing a program for controlling a computer device to function as each unit of the color processing device according to claim 1.   8. A computer-readable recording medium on which the program according to claim 7 is recorded.

Images