JP2007060421A - Image processor and its method, computer program, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To compute an optimum output signal by performing compression within a color area of output equipment when the spectrum reflectivity of an input is not within the color area of the output equipment. <P>SOLUTION: An image output device inputs spectral reflectivity data acquired by input equipment capable of acquiring the spectral reflectivity data as an input signal, and converts them into an output signal of the output equipment. The output device has a means which has a Look-Up-Table of the input signal and an output signal, and converts the input signal into the output signal; a means for deciding whether or not the input signal is outside the color reproduction range of the output equipment; and a means for switching the Look-Up-Table to a table having a smaller number of dimensions to perform compression within the color reproduction range, when the input signal is outside the color reproduction range of the output equipment. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and method, a computer program, and a storage medium.

  Conventionally, when image data obtained from an input device such as a digital camera or scanner is faithfully reproduced by a display, projector, printer, etc., based on the three primary color theory, conditional color is used, and input tristimulus values and output Colorimetric color reproduction that matches tristimulus values has been performed.

  For example, there are an XYZ color system and a CIE-Lab color system that are generally used as a space in which tristimulus values are quantified. In these, it is possible to match colors faithfully by matching the XYZ tristimulus values and L * a * b * values of the original image and the reproduced image under a certain illumination light source. However, since the XYZ tristimulus values and L * a * b * values are affected by the illumination light source, the XYZ tristimulus values and L * a * b * values of the original image and output image match under a certain illumination light source. Even so, there may be cases where the values are completely different under different illumination light sources.

  For example, the two spectral reflectances (solid line and broken line) shown in FIG. 14 are the same L * a * b * values (L) under the illumination of the light D50 light source of the CIE auxiliary standard, although the spectral reflectances are different from each other. * = 66, a * = 16, b * = 18), which are perceived as the same color. However, under the illumination of the CIE auxiliary standard light A light source, different L * a * b * values (L * = 68, a * = 19, b * = 20 solid line), (L * = 68, a * = 16 , b * = 22 broken line), which is perceived as a different color.

  In this way, in colorimetric color reproduction, the color reproduction of the original image is performed on the assumption that the illumination light source to be observed is fixed. Therefore, the color matching relationship between the original image and the reproduction image is established under any illumination light source. There is a problem of disappearing.

  Therefore, in recent years, a more faithful color reproduction method is desired in the field of color management and the like. Specifically, spectral color reproduction that reproduces spectral reflectance, which is color information unique to an object that does not include illumination light source information, and makes the original image and the reproduced image look the same even under any illumination light source. It is desired.

  However, when the conventional three primary colors of RGB or CMY are used, it is premised on colorimetric color reproduction, so even if the tristimulus values match, the spectral reflectances do not match. Thus, it is known that spectral color reproduction can be achieved by controlling spectral reflectance using multiple primary colors (three or more primary colors) having independent spectral reflectances.

  In recent years, research on multiband cameras has been actively conducted as an input device that can acquire the spectral reflectance of a subject. A multi-band camera is a camera that enables acquisition of a large amount of subject information by shooting with a plurality of bands of four or more bands, unlike a conventional three-band input system of R, G, and B. By using this multiband camera, it is possible to acquire a spectral image having a spectral reflectance for each pixel.

  As output, research on multi-primary color displays using two projectors has been conducted. Furthermore, even in an ink jet printer, it is possible to reproduce the original spectral reflectance with high accuracy by using multi-primary color inks having independent spectral reflectances in addition to the conventional C, M, Y, and K inks. It is known.

  By using the above-mentioned multiband camera and printer or projector, it is possible to construct a spectral color reproduction system that acquires a spectral image of a subject and faithfully reproduces it.

  Here, in order to construct a spectral color reproduction system, the output device must convert the spectral reflectance acquired by the input device into an output signal that faithfully reproduces. However, in general, it is difficult to completely reproduce the input spectral reflectance except for a special case, and an error always occurs there. It is necessary to reduce this error and convert it into an output signal that reproduces the input spectral reflectance.

  As a method of performing spectral color reproduction, when reproducing the spectral reflectance of the target color, the n-color LED element emits light, and the error between the spectral reflectance of the target color and the spectral reflectance of the mixed light is minimized. The 1st method of determining the color mixture ratio of the LED element to make is proposed (refer patent document 1).

  A second method for performing spectral color reproduction using an ink jet printer has also been proposed (see Patent Document 2). In this proposal, principal component analysis is performed on a spectral image, and the spectral image is converted into an addition rate of the principal component vector (hereinafter referred to as a principal component coefficient) using the calculated principal component vector. Furthermore, the principal component coefficient is converted into an output signal (hereinafter referred to as ink amount) to the ink jet printer to reproduce the input spectral image.

  However, if the input spectral reflectance is outside the reproducible color reproduction range (hereinafter referred to as the color gamut) of the output device, it must be replaced with a color that is closest to the input spectral reflectance within the color gamut of the output device. In the case of the first method, since the combination of output signals closest to the input spectral reflectance must be calculated from all the combinations of output signals, there is a problem that it takes cost and time to calculate. .

  In the second method, the principal component coefficients arranged at equal intervals and the ink amount are associated with each other, and the ink amount is calculated from the input principal component coefficient. It is not specified how to process.

Furthermore, a method of compressing to the output color gamut when the input color gamut and the output color gamut are different in colorimetric color reproduction has been proposed (see Patent Document 3). Is not expected. Moreover, even if the method is extended to multi-dimensions, for example, the principal component coefficients of the input are compressed in the space in the multi-dimensional space of the principal component coefficients, each principal component vector as an axis is an independent axis. . And since the amount of information to hold differs, a simple compression method cannot compress the input spectral reflectance with high accuracy.
JP 2003-143417 A JP 2002-254708 A JP 2004-32140 A

  As described above, conventionally, when the input spectral reflectance is out of the color gamut of the output device, it is impossible to calculate the optimum output signal by compressing the input spectral reflectance into the color gamut of the output device.

  The present invention has been made in view of such a problem, and an object of the present invention is to calculate an optimal output signal by compressing the input device in the color gamut of the output device when the spectral reflectance of the input is out of the color gamut of the output device. And

As a means for achieving the above object, the present invention comprises the following arrangement. An image processing apparatus according to the present invention is an image output apparatus that converts spectral reflectance data acquired by an input device capable of acquiring spectral reflectance data as an input signal into an output signal in an output device. Means for converting the input signal into the output signal; means for determining whether the input signal is outside the color reproduction range of the output device; and And means for switching the Look-Up-Table to one having a smaller number of dimensions and compressing it within the color reproduction range when it is outside the color reproduction range of the output device.
An image processing method according to the present invention is an image output apparatus that converts spectral reflectance data acquired by an input device capable of acquiring spectral reflectance data as an input signal into an output signal in an output device. Means for converting the input signal into the output signal; means for determining whether the input signal is outside the color reproduction range of the output device; and And means for switching the Look-Up-Table to one having a smaller number of dimensions and compressing it within the color reproduction range when it is outside the color reproduction range of the output device.

  According to the present invention, when the input spectral reflectance is out of the color gamut of the output device, it can be compressed within the color gamut of the output device, and an optimal output signal can be calculated.

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

[First Embodiment]
Embodiments according to the present invention will be described below in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a spectral color reproduction apparatus corresponding to the first embodiment of the present invention. In FIG. 1, reference numeral 1 denotes an entire spectral color reproduction apparatus corresponding to an embodiment of the present invention. Reference numeral 2 denotes a spectral reflectance measuring device that measures the spectral reflectance of an object such as a spectral radiance meter. Reference numeral 3 denotes a spectral reflectance measuring unit for controlling the spectral reflectance measuring apparatus 2 and incorporates spectral reflectance data into the spectral color reproduction apparatus 1. Reference numeral 4 denotes a conversion LUT creation unit that calculates a principal component vector and a conversion LUT from spectral reflectance data measured by the spectral reflectance measurement unit 3 using the spectral reflectance measurement device 2. A principal component vector storage unit 5 stores the principal component vectors calculated by the conversion LUT creation unit 4. A conversion LUT storage unit 6 stores the conversion LUT calculated by the conversion LUT creation unit 4.

  Reference numeral 7 denotes an image input device for taking a spectral image of an object such as a multiband camera. Reference numeral 8 denotes an image input unit for inputting a spectral image photographed using the image input device 7 into the spectral color reproduction device 1. Reference numeral 9 denotes an input image storage unit that stores a spectral image acquired by the image input device 7 and input into the spectral color reproduction device 1 via the image input unit 8. 10 uses the principal component vector stored in the principal component vector storage unit 5 and the conversion LUT stored in the conversion LUT storage unit 6 to ink the spectral image stored in the input image storage unit 9. It is an ink amount calculation unit for converting into an ink amount image as amount data.

  Reference numeral 11 denotes an output image storage unit that stores an ink amount image obtained by the conversion in the ink amount calculation unit 10. An image display unit 12 performs control for displaying the images stored in the input image storage unit 9 and the output image storage unit 11 on the image display device 13. Reference numeral 13 denotes an image display device for displaying an image according to the control of the image display unit 12, and is composed of a CRT, LCD, or the like. Reference numeral 14 denotes an image output unit for controlling the image output device 15 to output the ink amount image stored in the output image storage unit 11. Reference numeral 15 denotes an image output device that prints out an ink amount image in accordance with the control of the image output unit 14, and is configured by an inkjet printer or the like. Reference numeral 16 denotes a setting storage unit that stores setting information for designating a conversion method and the like in the conversion LUT and the ink amount calculation unit 10.

<User interface>
With reference to FIG. 9, an example of a user interface in the first embodiment will be described. 9 represents the entire display screen, and the screen 900 is displayed on the image display device 13. In the screen 900, reference numeral 901 denotes an input image display unit for displaying a spectral image stored in the input image storage unit 9. Reference numeral 902 denotes an ink amount image display unit 902 for displaying the ink amount image stored in the output image storage unit 11. Reference numeral 903 denotes an output image display unit for displaying an output image. Reference numeral 904 denotes a cursor for designating an arbitrary position in the screen 900, and the operation of the cursor is controlled by the user of the spectral color reproduction device 1.

  Reference numeral 905 denotes a setting information unit that displays setting information. Reference numeral 906 denotes a light source information selection unit 906 for selecting light source information from the setting information. Reference numeral 907 denotes a used dimension number setting unit for setting the used dimension number in the setting information. Reference numeral 908 denotes a search dimension number setting unit for setting the search dimension number in the setting information. Reference numeral 909 denotes an evaluation function setting unit for setting an evaluation function in the setting information. Reference numeral 910 denotes a table creation button for accepting a conversion LUT creation instruction from the user.

  Reference numeral 911 denotes an image reading button for accepting a spectroscopic image acquisition instruction using the image input device 7 from the user. Reference numeral 912 denotes a conversion button for receiving an instruction to convert a spectral image to generate an ink amount image. Reference numeral 913 denotes a print button for receiving a print output instruction for the ink amount image stored in the output image storage unit 11. Reference numeral 914 denotes a spectral reflectance display unit that displays the spectral reflectance calculated from the input image and the spectral reflectance calculated from the output image.

<Overall process flow>
The overall process flow corresponding to the present embodiment will be described below. FIG. 2 is a flowchart corresponding to an example of the operation flow of the spectral color reproduction device 1 corresponding to the present embodiment. FIG. 9 is a diagram illustrating an example of a user interface used when performing color reproduction. In the following description, six color inks of C, M, Y, K, R, and G are used as an example, and the spectral reflectance is obtained by sampling the visible light region 380 to 730 nm every 10 nm. Dimensional data shall be used.

  In FIG. 2, first, in step S201, it is determined whether or not the table creation button 910 has been pressed. If the table creation button is pressed (“YES” in step S201), the process proceeds to step S202. In step S202, parameters set in the light source information selection unit 906, the used dimension number setting unit 907, the search dimension number setting unit 908, and the evaluation function setting unit 909 in the setting information unit 905 are acquired and stored in the setting information storage. Store in unit 16.

  In step S203, the conversion LUT creation unit 4 calculates the principal component vector and the conversion LUT, and stores them in the principal component vector storage unit 5 and the conversion LUT storage unit 6, respectively. The processing in the conversion LUT creation unit 4 will be described later with reference to FIGS. In step S204, it is determined whether the image reading button 911 has been pressed. If pressed (“YES” in step S204), the process proceeds to step S205. In step S <b> 205, the image input unit 8 acquires a spectral image from the image input device 7 and stores it in the input image storage unit 9. In step S <b> 206, the spectral image stored in the input image storage unit 9 is displayed on the input image display unit 901. More specifically, an XYZ tristimulus value is calculated for each pixel of the spectral image stored in the input image storage unit 9. Next, the calculated XYZ tristimulus values are converted into device RGB values which are color space data that can be displayed by the image display device 13 using an ICC profile or the like. Finally, the calculated RGB values are displayed on the input image display unit 901 using the image display device 13 in the image display unit 9.

  In step S207, it is determined whether the conversion button 912 has been pressed. If it is pressed (“YES” in step S207), the process proceeds to step S208. In step S <b> 208, the ink amount calculation unit 10 calculates an ink amount image and stores it in the output image storage unit 11. This is performed using the conversion LUT stored in the conversion LUT storage unit 6 and the spectral image stored in the input image storage unit 9 based on the setting information stored in the setting information storage unit 16. .

  In subsequent step S209, the ink amount image stored in the output image storage unit 11 is displayed on the ink amount image display unit 902. In step S210, based on the light source information stored in the setting information storage unit 16, the image display unit 12 converts the ink amount image stored in the output image storage unit 11 into an RGB image of the device, and the image display device. 13 is displayed. In step S211, it is determined whether the output image display unit 903 has been clicked. If it is clicked (“YES” in step S211), the process proceeds to step S212, and if it is not clicked (“NO” in step S211), the process proceeds to step S213.

  In step S212, the clicked position information is acquired, and the spectral reflectance corresponding to the position information is displayed on the spectral reflectance display unit 914 from the input image storage unit 9 and the output image storage unit 11, respectively. In step S213, it is determined whether the print button 913 has been pressed. If it is pressed (“YES” in step S213), the process proceeds to step S214. In step S <b> 214, the image output unit 14 outputs the ink amount image stored in the output image storage unit 11 using the image output device 15.

<Conversion LUT creation unit>
FIG. 3 is a block diagram showing an example of a detailed configuration of the conversion LUT creation unit 4 in the spectral color reproduction device 1 corresponding to the embodiment of the present invention.

  In FIG. 3, 301 corresponds to the conversion LUT creation unit 3. A sample image spectral reflectance storage unit 302 stores the spectral reflectance of the sample image input from the spectral reflectance measurement unit 3 of FIG. A principal component vector calculation unit 303 performs principal component analysis on the spectral reflectance of the sample image stored in the sample image spectral reflectance storage unit 302 and calculates a principal component vector. A principal component coefficient calculation unit 304 calculates a principal component coefficient from the principal component vector calculated by the principal component vector calculation unit 303 and the spectral reflectance of the sample image stored in the sample image spectral reflectance storage unit 302. is there. A sample image ink amount storage unit 305 stores the ink amount of the sample image.

  Next, a detailed flow of the conversion LUT creation processing (corresponding to step S203 in FIG. 2) in the conversion LUT creation unit 4 will be described with reference to the flowchart of FIG.

In FIG. 4, first, in step S <b> 401, a sample image is output by the image output unit 14 using the image output device 15, and an ink amount image of the output sample is stored in the sample image ink amount storage unit 305. The sample image here, to C, M, Y, K, R, and m ^6-color patch of each ink amount was changed to m stages of G. Increasing the value of m increases the interpolation accuracy, but the number of data increases enormously. Therefore, in the present embodiment, 15625 colors of m = 5 are used as sample images as an example.

  Next, in step S <b> 402, the spectral reflectance measurement unit 3 acquires the spectral reflectance of the sample image using the spectral reflectance measurement device 2 and stores it in the sample image spectral reflectance storage unit 302. In subsequent step S403, the principal component vector calculation unit 303 performs principal component analysis on the spectral reflectance stored in the sample image spectral reflectance storage unit 302 to calculate a principal component vector. The calculated principal component vector is stored in the principal component vector storage unit 5.

  In step S404, based on the setting information stored in the setting information storage unit 16, the principal component coefficient calculation unit 304 changes the number of used dimensions of the principal component vector stored in the principal component vector storage unit 5. Let In addition, a principal component coefficient for each use dimension number of the principal component vector is calculated from the spectral reflectance data stored in the sample image spectral reflectance storage unit 302. As a result, a conversion LUT in which the ink amount data of the sample image stored in the sample image ink amount storage unit 305 is associated with the calculated principal component coefficient for each number of used dimensions is created. That is, the conversion LUT is created for each number of principal component coefficients corresponding to the number of used dimensions.

  Finally, in step S405, the converted LUT data created in step S404 is stored in the converted LUT data storage unit 17.

<Calculation of principal component coefficients>
Here, the principal component coefficient calculation processing for each number of used dimensions in step S404 in FIG. 4 will be described in more detail.

First, the principal component vector calculation unit 303 performs principal component analysis on the spectral reflectances stored in the sample image spectral reflectance storage unit 302, and calculates the calculated principal component vectors.
Assuming that (i = 1 to 36), the spectral reflectance is expressed by the following equation (1) using the principal component coefficients a _i (i = 1 to 36).
... (1)
Here, m is an average value vector. The principal component vector obtained by principal component analysis has information on the entire population in the first principal component, and detailed information on the population as it becomes second and third, so as the number of dimensions increases. Less information. Therefore, even if the principal component vectors up to the lower order are used, it is possible to highly approximate the input spectral reflectance. As an example, when the input spectral reflectance is approximated using up to the sixth principal component vector, it is expressed as the following equation (2).
... (2)
here,
Then,
... (3)
The principal component coefficient a can be obtained from the following equation (4).
... (4)
Here, by changing the number of dimensions used i = 1 to 36, it is possible to calculate principal component coefficients for each number of used dimensions.

  In the present embodiment, the number of used dimensions setting unit 907 changes the number of dimensions up to the number of used dimensions specified by the user. As an example, if 6 is set in the used dimension number setting unit 907, i = 1 to 6 is changed.

<Ink amount calculation unit>
FIG. 5 is a block diagram showing an example of a detailed configuration of the ink amount calculation unit 10 in the spectral color reproduction device 1 corresponding to the embodiment of the present invention.

  In FIG. 5, 501 corresponds to the ink amount calculation unit 10. Reference numeral 502 denotes a spectral image stored in the input image storage unit 9 by using the conversion LUT stored in the conversion LUT storage unit 6 and the principal component vector stored in the principal component vector storage unit 5. It is a principal component coefficient calculation unit for converting into component coefficients. 503 uses the conversion LUT stored in the conversion LUT storage unit 6 to search whether the input spectral reflectance is within the color gamut of the output device, and calculates the ink amount corresponding to the input spectral reflectance. An ink amount search unit. Reference numeral 504 denotes a compression unit that compresses the spectral reflectance into the color gamut of the output device when the input spectral reflectance is outside the color gamut of the output device. Reference numeral 505 denotes a selection unit that selects an optimum ink amount from the calculated ink amounts using the conversion LUT stored in the conversion LUT storage unit 6.

  Next, a detailed flow of process conversion (corresponding to step S208 in FIG. 2) for converting a spectral image into an ink amount will be described with reference to the flowchart in FIG.

  First, in S601, the search dimension number N is set to a setting value received from the user via the search dimension number setting unit 908 in FIG. In subsequent step S602, the principal component coefficient calculation unit 502, for each pixel of the spectral image stored in the input image storage unit 9, the principal component vector data stored in the principal component vector storage unit 5 and the above-described component data. An N-dimensional principal component coefficient is calculated using Equation (4). In step S603, the ink amount search unit 503 searches for the ink amount corresponding to the input principal component coefficient using the N-dimensional conversion LUT stored in the conversion LUT storage unit 6.

  If it is determined in step S604 that the N-dimensional conversion LUT is used and the N-dimensional LUT is not used ("YES" in step S604), the process proceeds to step S606. On the other hand, if it is determined that the ink is in the N-dimensional LUT and the corresponding ink amount is found (“NO” in step S604), the process proceeds to step S605. In step S605, N = N-1 is calculated in the compression unit 504, and the process returns to step S602 again.

  The above processing is repeated until the input N-dimensional principal component coefficient is present in the N-dimensional conversion LUT. In step S606, the ink selection unit 505 selects an optimal ink amount and stores the calculated ink amount image in the output image storage unit 11.

<Ink amount search unit>
Hereinafter, the ink amount searching process of the ink amount searching unit 503 in step S603 in FIG. 6 will be described in more detail.

  First, the N-dimensional principal component coefficient calculated by the principal component coefficient calculation unit 502 in step S602 is input to the ink amount search unit 503. When the principal component coefficients input in this way are points other than the grid of the conversion LUT, the ink amount search unit 503 calculates an interpolation value corresponding to the input by interpolation using neighboring points.

As an example, here, N = 6 as the number of inks. A general interpolation value calculation method will be described. First, input point
Then, using the neighboring points a _{0 to} a ₆ and the coefficients b _{1 to} b ₆ , they can be expressed as
... (6)
Here, in order for a to exist in the region generated by the neighboring points a _{0 to} a ₆ ,
... (7)
Is a necessary and sufficient condition.

Therefore, the coefficients b _{1 to} b ₆ are
... (8)
Can be calculated as follows. Here, T represents transposition.

  Based on the above, when a principal component coefficient is input as an input using Expression (7) and Expression (8), the corresponding coefficient b is calculated. Next, if the input is considered as the ink amount corresponding to the principal component coefficient using the calculated coefficient b, the ink amount after interpolation can be calculated by using Equation (6) in the same manner.

<Ink selection part>
Next, the ink amount selection process in the ink selection unit 505 in step S606 will be described in detail with reference to the flowchart of FIG.

  In the ink amount calculation processing using the N-dimensional conversion LUT in step S603 of FIG. 6, if no interpolation point is found, N = N−1 is set by the compression unit 504, and the number of search dimensions is calculated until the ink amount is calculated. Is reduced. As a result, (dimension number of input principal component coefficient) <(number of output ink), and there are a plurality of combinations of ink amounts representing the same principal component coefficient. Therefore, an optimum ink amount must be selected from a combination of a plurality of ink amounts.

First, in step S701 in FIG. 7, using the conversion LUT stored in the conversion LUT storage unit 6, the calculated ink amount is converted into a principal component system number using the above equation (6). Next, in step S702, the calculated principal component coefficients and principal component vector data stored in the principal component vector storage unit 5 are substituted into equation (3), and converted to spectral reflectance O _predict (λ). . Finally, in step S703, an evaluation value E with the corresponding spectral reflectance O _predict (λ) of the spectral image stored in the input image storage unit 9 using the evaluation function selected by the evaluation function setting unit 909. The solution that minimizes is calculated as the optimal solution.

<Evaluation function>
As an example of calculating the evaluation value E,
... (9)
A method of minimizing the sum of errors for each wavelength may be adopted. Further, using a weighting function W (λ) that places importance on a region where the influence of human vision is large, as disclosed in JP-A-2003-333355,
... (10)
A calculation method can also be adopted. Further, there is a method in which the light source setting unit 906 calculates tristimulus values under the light source for which setting is received from the user, and calculates the color difference ΔE under the light source as an error value. The evaluation function setting unit 909 in FIG. 9 can accept selection of any one of these methods from the user.

<Display output image>
Here, the process in step S210 of FIG. 2 will be described in detail. First, using the conversion LUT storage unit 6, the ink amount of each pixel of the ink amount image stored in the output image storage unit 11 is converted into a principal component coefficient, and further spectral reflection is performed using the above equation (3). Convert to rate and acquire spectral image. Next, as in the case of displaying a spectral image, the image display device 13 converts it into device RGB which is color space data that can be displayed. Finally, the calculated RGB values are displayed on the output image display unit 903 by using the image display device 13 in the image display unit 12.

<Display of spectral reflectance>
Next, the process in step S212 of FIG. 2 will be described in detail with reference to the flowchart of FIG.

  In FIG. 8, in step S <b> 801, position information clicked by the user using the cursor 904 in the output image display unit 903 is acquired. In the subsequent step S802, the spectral reflectance at the position indicated by the position information is acquired from the spectral image stored in the input image storage unit 9. In step S803, the ink amount at the position indicated by the position information is acquired from the ink amount image stored in the output image storage unit 11. In step S804, a principal component coefficient corresponding to the ink amount is calculated using the conversion LUT stored in the conversion LUT storage unit 6, and the spectral reflectance is calculated using the above-described equation (3). . Finally, in step S805, the spectral reflectance calculated from the input image and the spectral reflectance calculated from the output image are respectively displayed on the spectral reflectance display unit 914 (input: solid line, output: broken line).

<Effect of the first embodiment>
As described above, according to the present invention corresponding to the present embodiment, even when the spectral reflectance of the input is out of the color gamut of the output device in the spectral color reproduction processing for converting the input spectral image into the output signal. Thus, it is possible to convert the output signal into an optimum output signal or ink amount within the color gamut of the output device.

[Second Embodiment]
Below, the 2nd Embodiment of this invention is described in detail with reference to drawings.

  FIG. 10 is a block diagram showing an example of the configuration of the spectral color reproduction device corresponding to the present embodiment. In FIG. 10, reference numeral 1001 denotes a spectral color reproduction apparatus corresponding to this embodiment. Reference numeral 1002 denotes a sample image storage unit that stores spectral reflectance data of a sample image and an ink amount image as ink amount data. Reference numeral 1003 denotes a principal component vector storage unit that stores principal component vectors. A conversion LUT creation unit 1004 creates a conversion LUT. The creation of the conversion LUT is stored in the spectral reflectance, ink amount, and principal component vector storage unit 1003 of the sample image stored in the sample image storage unit 1002 based on the setting information stored in the setting information storage unit 1013. Using the principal component vector.

  A conversion LUT storage unit 1005 stores the conversion LUT created by the conversion LUT creation unit 1004. Reference numeral 1006 denotes an input image storage unit that stores an input spectral image. Reference numeral 1007 denotes an input spectral image stored in the input image storage unit 1006 using the principal component vector stored in the principal component vector storage unit 1003 and the conversion LUT stored in the conversion LUT storage unit 1005. Is an ink amount calculation unit that converts the ink amount into an ink amount. Reference numeral 1008 denotes an output image storage unit that stores the ink amount image generated by the ink amount calculation unit 1007.

  Reference numeral 1009 denotes an image display unit for displaying images stored in the input image storage unit 1006 and the output image storage unit 1008. Reference numeral 1010 denotes an image display device, which includes a CRT, an LCD, or the like. Reference numeral 1011 denotes an image output unit that outputs the ink amount image stored in the output image storage unit 1008. Reference numeral 1012 denotes an image output apparatus, which is composed of, for example, an ink jet printer. A setting information storage unit 1013 stores setting information for designating a conversion method and the like in the conversion LUT creation unit 1004 and the ink amount calculation unit 1007.

<User interface>
FIG. 13 is a diagram illustrating an example of a user interface used when performing color reproduction.

  Reference numeral 1300 in FIG. 13 represents the entire display screen, and the screen 1300 is displayed on the image display device 13. In the screen 1300, reference numeral 1301 denotes an edit box, which shows the file name, storage location, and the like of sample data. Reference numeral 1302 denotes a sample data reading button, which accepts an instruction to read sample data shown in the edit box 1301 from the user. Reference numeral 1303 denotes an edit box, which shows the file name, storage location, etc. of the principal component vector file. Reference numeral 1304 denotes a principal component vector reading button, which accepts an instruction to read a principal component vector file shown in the edit box 1303 from the user. An edit box 1305 indicates the file name, storage location, etc. of the input image data. Reference numeral 1306 denotes an input image reading button, which accepts an input image data reading instruction shown in the edit box 1305 from the user.

  Reference numeral 1307 denotes a principal component vector display unit. Reference numerals 1308 to 1313, 1314, and 1315 correspond to 905 to 910, 912, and 913 in FIG. Reference numeral 1316 denotes an input image display unit for displaying a spectral image stored in the input image storage unit 1006. Reference numeral 1317 denotes an output image display unit for displaying an output image.

<Overall processing>
Next, the flow of processing in the spectral color reproduction device 1001 corresponding to this embodiment will be described with reference to the flowchart of FIG. As in the first embodiment, for simplicity of explanation in this embodiment, inks of six colors C, M, Y, K, R, and G are used, and the spectral reflectance is in the visible light region. 36-dimensional data obtained by sampling 380 to 730 nm every 10 nm will be used.

  First, in S1101, it is determined whether the sample data reading button 1302 has been pressed. If it is pressed (“YES” in step S1101), the process proceeds to step S1102. In step S1102, the sample image spectral reflectance data and the ink amount image described in the edit box 1301 are stored in the sample image storage unit 1002. In step S1103, it is determined whether the principal component vector reading button 1304 has been pressed. If it is pressed (“YES” in step S1103), the process proceeds to step S1104. In step S 1104, the principal component vector data described in edit box 1303 is stored in principal component vector storage section 1003.

  In step S1105, the principal component vector data stored in the principal component vector storage unit 1003 is displayed on the principal component vector display unit 1307. In step S1106, it is determined whether the input image reading button 1306 has been pressed. If it has been pressed (“YES” in step S1106), the process proceeds to step S1107. In step S 1107, the input image data described in the edit box 1305 is stored in the input image storage unit 1306. In step S 1108, the input image stored in the input image storage unit 1006 is displayed on the input image display unit 1316.

  In a succeeding step S1109, it is determined whether or not the table creation button 1313 is pressed. If it is pressed (“YES” in step S1109), the process proceeds to step S1110. In step S1110, parameters set using the light source information selection unit 1309, the used dimension number setting unit 1310, the search dimension number setting unit 1311, and the evaluation function setting unit 1312 are acquired from the setting information unit 1308 and set. The information is stored in the information storage unit 1013.

  In step S1111, the conversion LUT generation unit 1005 generates a conversion LUT. This creation is performed based on the setting information in the setting information storage unit 1013, using the principal component vector in the principal component vector storage unit 1003, the spectral reflectance data of the sample image in the sample image storage unit 1002, and the ink amount image. . The conversion LUT created here is stored in the conversion LUT storage unit 1005. In step S1112, it is determined whether the image conversion button 1314 has been pressed. If it is pressed (“YES” in step S1112), the process proceeds to step S1113.

  In step S <b> 1113, the ink amount calculation unit 1007 converts the input spectral image stored in the input image storage unit 1006 into an ink amount image and stores it in the output image storage unit 1008. In step S 1114, the ink amount image stored in the output image storage unit 1008 is displayed on the output image display unit 1317. In a succeeding step S1115, it is determined whether or not the print button 1315 is pressed. If it is pressed (“YES” in step S1115), the process proceeds to step S1116. Finally, in step S1116, the ink amount image stored in the output image storage unit 1008 is output by the image output device 1012 via the image output unit 1011.

<Conversion LUT creation unit>
With reference to FIG. 12, an example of a detailed configuration of the conversion LUT creation unit 1004 in the spectral color reproduction device 1001 corresponding to the present embodiment will be described.

  In FIG. 12, reference numeral 1201 denotes a conversion LUT creation unit. 1202 calculates a principal component coefficient based on the spectral reflectance data and the ink amount image of the sample image stored in the sample image storage unit 1002 and the principal component vector stored in the principal component vector storage unit 1003. The principal component coefficient calculation unit.

  Since the conversion LUT creation unit 1004 in this embodiment reads the principal component vector from an external file or the like, the principal component vector creation processing in the first embodiment (processing corresponding to steps S401 to S403 in FIG. 4) is unnecessary. It becomes. However, other processing contents are the same as those in the first embodiment.

<Effects of Second Embodiment>
As described above, by using this embodiment, the sample image, the principal component vector, and the input spectral image can be captured from the external file. Arbitrary principal component vectors can be captured, so image processing is performed using principal component vectors of input images, principal components of output images from output devices, and standard principal component vectors Is possible.

  Similarly to the first embodiment, in the spectral color reproduction process for converting an input spectral image into an output signal, even if the input spectral reflectance is outside the color gamut of the output device, the color gamut of the output device. It is possible to convert to an optimum output signal.

<Spectral image input>
In the above embodiment, the image input device 7 is used to input a spectral image. However, a spectral image may be acquired via a communication network, or a multiband image may be used instead of a spectral image. Also, a compressed image format may be used. That is, it is only necessary to obtain a spectral image by conversion or decompression.

<Wavelength calculation range and sampling interval>
In the above embodiment, the spectral reflectance does not have to be a limited wavelength range (for example, 380 to 730 nm) and an interval (for example, 10 nm). You can widen or narrow the sampling interval. On the other hand, the calculation range can be reduced by narrowing the wavelength range and widening the sampling interval. That is, the wavelength range and sampling interval may be changed according to the accuracy and calculation amount desired by the user.

<Number of inks>
In the above embodiment, the number of inks is six. However, the conventional three colors of RGB and CMY may be used, or six or more colors may be used. In other words, any number of colors can be employed when implementing the present invention.

<Sample image>
In the above embodiment, h, for example, the case where the ink amount of the sample image is changed in five stages has been described. However, the stage of change is not limited to five stages, and any number of stages may be used. In the first embodiment, a sample image is created during processing. However, as in the second embodiment, an external file in which the spectral reflectance of the sample image has been measured in advance may be read. Absent.

<Principal component vector>
In the above description, the second embodiment describes the case where the principal component vector of the external file is read. However, the principal component vector to be read may be a principal component vector calculated from a sample image of the output device, a principal component vector calculated from an input spectral image, or a standard principal component vector. It is desirable for the user to select according to the accuracy.

<Calculation method of principal component coefficient>
In the above embodiment, it is also possible to calculate the principal component coefficient by attaching importance to an appropriate region of the spectral reflectance by using a weight function for calculating the principal component coefficient. At that time, the calculation method of the principal component coefficient can be expressed as the following equation (5).
... (5)
here,
It is.

<Optimization of ink amount>
In the above embodiment, the optimum ink amount is selected using the error value for the ink amount calculated after reducing the search dimension, but by operating each ink amount around the calculated ink amount. It is also possible to calculate a more accurate solution.

<Number of dimensions used>
In the above embodiment, the number of dimensions used is six. This is because the number of dimensions used is greater than the number of inks for creating the conversion LUT because the ink has six colors. In addition, since the number of dimensions used inevitably increases when the number of inks is increased, the improvement in reproduction accuracy is expected.

<Number of search dimensions>
In the above embodiment, since there are six ink colors, the number of search dimensions is six. However, the number of search dimensions can be changed according to the application. For example, when the search is started in 6 dimensions, a high reproduction accuracy can be expected, but it is determined that the conversion LUT is out of the color gamut and the conversion LUT is switched again to search again. There is a risk that it will take. On the other hand, when the search is started in a low dimension such as three dimensions, the reproduction accuracy is lowered, but it is possible to determine a large number of inputs within the color gamut of the conversion LUT and calculate a solution at high speed.

<Evaluation function>
In the above embodiment, RMS, weighted RMS, and color difference ΔE are used as evaluation functions for selecting an optimum ink amount. However, CIEΔE94 color difference may be used, or a correlation coefficient between two spectral reflectances may be used. good. That is, anything that evaluates an error between the original spectral reflectance and the reproduced spectral reflectance is acceptable.

<User interface>
In the above embodiment, an example of the user interface is shown in FIGS. 9 and 13, but it goes without saying that the user interface is not limited to these forms. For example, in FIGS. 9 and 13, the user selects a desired light source in the light source selection unit 906 and the light source selection unit 1309, but a file in which light source data is previously described may be read. Further, although the spectral reflectance display unit 914 in FIG. 9 displays the spectral reflectance, it may be an XYZ tristimulus value or a CIE-L * a * b * value.

  Further, in FIG. 13, the user reads a file in which principal component vector data is described. However, the user may select a format selected from a list of principal component vectors stored in advance. In the evaluation function setting units 909 and 1312 in FIGS. 9 and 13, the user selects the evaluation function from the check boxes. However, the user may specify the evaluation function by an external file or the like. . In other words, any user interface configuration that allows user-desired settings may be used.

  As described above, according to the present invention, in the spectral color reproduction process for converting an input spectral image into an output signal, even if the input spectral reflectance is out of the color gamut of the output device, the color gamut of the output device. It is possible to convert to an optimum output signal.

[Other Embodiments]
The present invention can be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), or a device composed of a single device (for example, a copier, a facsimile device, etc.) You may apply to.

  Another object of the present invention is to supply a storage medium storing software program codes for realizing the functions of the above-described embodiments to a system or apparatus, and the computer (or CPU or MPU) of the system or apparatus stores the storage medium. Needless to say, this can also be achieved by reading and executing the program code stored in. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

  As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, ROM, etc. are used. be able to.

  In addition, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (operating system) running on the computer based on the instruction of the program code. It goes without saying that a part of the actual processing is performed and the functions of the above-described embodiments are realized by the processing.

  Further, after the program code read from the storage medium is written to a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that a case where the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

It is a block diagram which shows an example of a structure of the spectral color reproduction apparatus corresponding to the 1st Embodiment of this invention. It is a flowchart which shows an example of the flow of a process in the spectral color reproduction apparatus 1 corresponding to the 1st Embodiment of this invention. It is a block diagram which shows an example of a structure of the conversion LUT preparation part 4 corresponding to the 1st Embodiment of this invention. It is a flowchart which shows an example of the flow of conversion LUT creation processing corresponding to the 1st Embodiment of this invention. FIG. 3 is a block diagram illustrating an example of a configuration of an ink amount calculation unit 10 corresponding to the first embodiment of the present invention. It is a flowchart which shows an example of the flow of the process corresponding to the 1st Embodiment of this invention which converts a spectral reflectance into an ink amount. 6 is a flowchart illustrating an example of a flow of processing for selecting an optimal ink amount corresponding to the first embodiment of the present invention. It is a flowchart which shows an example of the process for displaying a spectral reflectance corresponding to the 1st Embodiment of this invention. It is a figure which shows an example of the user interface corresponding to the 1st Embodiment of this invention. It is a block diagram which shows an example of a structure of the spectral color reproduction apparatus corresponding to the 2nd Embodiment of this invention. 10 is a flowchart showing a flow of processing in the spectral color reproduction device 1001 corresponding to the second embodiment of the present invention. It is a block diagram which shows an example of a structure of the conversion LUT preparation part 1004 corresponding to the 2nd Embodiment of this invention. It is a figure which shows an example of the user interface corresponding to the 2nd Embodiment of this invention. It is a figure which shows the example of two spectral reflectivities with which a conditional color match is realized under the light D50 light source of CIE auxiliary standard.

Claims (16)

An image processing apparatus that converts spectral reflectance data into output data in an image output device and causes output by the connected image output device,
Table storage means for storing a plurality of conversion tables for registering the coefficient when the spectral reflectance data is represented by a principal component vector and its coefficient and the output data in association with each other;
Conversion means for converting the spectral reflectance data into the output data based on any one of the plurality of conversion tables;
Of the plurality of conversion tables, the output data obtained by converting the spectral reflectance data by the conversion unit based on a conversion table corresponding to a predetermined number of the coefficients is color reproduction of the image output device. Determination means for determining whether or not out of range,
When the determination unit determines that the image output device is out of the color reproduction range, the conversion unit further performs the conversion based on a conversion table corresponding to a coefficient smaller than the predetermined number. An image processing apparatus. When a plurality of output data determined not to be out of the color reproduction range of the image output device is obtained by the determination unit,
The conversion unit calculates the coefficient for each of the plurality of output data based on a conversion table corresponding to each of the plurality of output data, and the spectral reflectance based on the calculated coefficient and the principal component vector 2. The image processing according to claim 1, wherein data is estimated, and output data that minimizes an error between the spectral reflectance data and the estimated spectral reflectance data is determined in the plurality of output data. apparatus. Control means for controlling the output of the image output device;
Obtaining means for obtaining spectral reflectance data of a predetermined image output by the control means controlling the image output device;
Table creation means for creating the conversion table based on spectral reflectance data of the predetermined image;
The image processing apparatus according to claim 1, further comprising: The table creation means includes
Principal component vector calculating means for calculating the principal component vector by principal component analysis from spectral reflectance data of the predetermined image;
Coefficient calculation means for calculating the coefficient from spectral reflectance data of the predetermined image and the principal component vector,
4. The image according to claim 3, wherein the conversion table is created by associating output data for outputting the predetermined image stored in a storage unit with a coefficient calculated by the coefficient calculation unit. Processing equipment.   The spectral reflectance data stored in the storage means for storing the spectral reflectance data of the predetermined image and the output data for outputting the predetermined image, and the principal component vector of the spectral reflectance data of the predetermined image are stored. Table creation means for calculating the coefficient based on the principal component vector stored in the principal component vector storage means and creating the conversion table by associating the output data stored in the storage means with the coefficient The image processing apparatus according to claim 1, further comprising:   The image processing apparatus according to claim 1, wherein the image output device is a printer having four or more primary colors, and the output data is ink amount data for the printer. A monitor or projector with four or more primary colors is further connected,
Coefficient conversion means for converting the ink amount data into the coefficient based on the conversion table;
Generating means for generating spectral reflectance data based on the coefficient obtained by the coefficient converting means and the principal component vector;
Color space conversion means for converting spectral reflectance data obtained by the generation means into color space data that can be displayed by the image output device;
The image processing apparatus according to claim 6, wherein the monitor or projector displays the color space data obtained by the color space conversion unit. A method of controlling an image processing apparatus that converts spectral reflectance data into output data in an image output device and causes output by the connected image output device,
When the spectral reflectance data is represented by a principal component vector and its coefficient, the spectral reflectance is based on one of the plurality of conversion tables for registering the coefficient in association with the output data. A conversion step of converting data into the output data;
Among the plurality of conversion tables, the output data obtained by converting the spectral reflectance data in the conversion step based on a conversion table corresponding to a predetermined number of the coefficients is color reproduction of the image output device. And a determination step of determining whether or not out of range,
When it is determined in the determination step that the image output device is out of the color reproduction range, the conversion step further performs the conversion based on a conversion table corresponding to a coefficient smaller than the predetermined number. A control method of the image processing apparatus. In the determination step, when a plurality of output data determined not to be out of the color reproduction range of the image output device,
In the conversion step, the coefficient is calculated for each of the plurality of output data based on a conversion table corresponding to each of the plurality of output data, and the spectral reflection is calculated based on the calculated coefficient and the principal component vector. 9. The output data according to claim 8, wherein rate data is estimated, and output data that minimizes an error between the spectral reflectance data and the estimated spectral reflectance data is determined in the plurality of output data. A method for controlling an image processing apparatus. A control step of controlling the output of the image output device;
An acquisition step of acquiring spectral reflectance data of a predetermined image output by controlling the image output device in the control step;
A table creating step for creating the conversion table based on spectral reflectance data of the predetermined image;
An image processing apparatus control method according to claim 8 or 9, further comprising: In the table creation process,
A principal component vector calculation step of calculating the principal component vector by principal component analysis from spectral reflectance data of the predetermined image;
A coefficient calculating step of calculating the coefficient from spectral reflectance data of the predetermined image and the principal component vector,
11. The conversion table is created by associating output data for outputting the predetermined image stored in a storage unit with a coefficient calculated by the coefficient calculation unit. Method for controlling the image processing apparatus.   The spectral reflectance data stored in the storage means for storing the spectral reflectance data of the predetermined image and the output data for outputting the predetermined image, and the principal component vector of the spectral reflectance data of the predetermined image are stored. A table creation step of calculating the coefficient based on the principal component vector stored in the principal component vector storage means and creating the conversion table by associating the output data stored in the storage means with the coefficient The image processing apparatus control method according to claim 8, further comprising:   13. The control of an image processing apparatus according to claim 8, wherein the image output device is a printer having four or more primary colors, and the output data is ink amount data for the printer. Method. The image processing apparatus is further connected to a monitor or projector having four or more primary colors,
The control method of the image processing apparatus is:
A coefficient conversion step of converting the ink amount data into the coefficient based on the conversion table;
Based on the coefficient obtained in the coefficient conversion step and the principal component vector, a generation step for generating spectral reflectance data;
A spectral space conversion step of converting spectral reflectance data obtained in the generation step into color space data that can be displayed by the image output device;
The method of controlling an image processing apparatus according to claim 13, wherein the monitor or projector displays the color space data obtained in the color space conversion step.   A computer program for causing a computer to execute the control method for an image processing apparatus according to claim 8.   A computer-readable storage medium storing the computer program according to claim 15.