JP2007251618A - Image processing apparatus, image processing method and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that it requires time and effort to realize a style with a sense of noise. <P>SOLUTION: An image processing apparatus converts first image data into second image data, by referring to a color conversion table defining the correspondence relation, for a representative gradation value, between the first image data in which the color of each of pixels is expressed in gradation by a first color specification space and second image data in which the color of each of pixels is expressed in gradation by a second color specification space. The apparatus is provided with a means for acquiring the first image data representing a predetermined image; a means for causing disturbance in a predetermined distributing rule, in executing processing of distributing the acquired first image data to any one of neighboring representative gradation values in the first color specification space, on the basis of the predetermined distributing rule; and a means for reading the second image data corresponding to the representative gradation value in the first specification space of a distribution destination, while referring to the color conversion table. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program that execute color conversion processing.

  Conventionally, as a part of processing when a desired image is output by an image output device, image data represented by a certain color space (for example, red (R), green (G), blue (B) The image data expressed by each gradation value) is converted into image data expressed by another color space (for example, ink data defining the recording amount for each ink type used in the printing apparatus) and output. Color conversion processing is performed. The above color conversion process refers to a color conversion table that defines the conversion relationship between image data in the color space before conversion and image data in the color space after conversion for each representative coordinate in the color space before conversion. It is common to do this.

Here, conventionally, input image data to be subjected to color conversion is distributed to neighboring coordinate points among the representative coordinate points that define the conversion relationship in the color conversion table (this distribution process is referred to as pre-conversion). In other words, the color conversion may be realized by reading the converted image data associated with the coordinate point of the distribution destination (see Patent Document 1).
JP 2000-209435 A

  When a desired image is output by an image output device such as a printer, the user may desire to realize a style with irregular disturbances (style with noise) on the output image. However, in the conventional color conversion process, since the image data is only converted from one color space to another color space, the noise-like style cannot be realized on the image. In addition, the user has displayed image data before color conversion on a monitor using retouch software or the like and applied a noise filter to the image data to realize a favorite style. However, such work requires a lot of work time and operation skill for the user.

  The present invention has been made in view of the above problems, and an object thereof is to provide an image processing apparatus, an image processing method, and an image processing program capable of appropriately and easily realizing a style desired by a user in an output image. To do.

  In order to achieve the above object, in the image processing apparatus according to the present invention, the first image data in which the color of each pixel is represented by gradation in the first color space and the color of each pixel in the second color space are represented by gradation. The first image data is converted into the second image data with reference to a color conversion table that defines the correspondence relationship with the expressed second image data for the representative gradation value. The image data acquisition unit acquires first image data representing a predetermined image. Then, before performing the conversion with reference to the color conversion table for the acquired first image data, the image data distribution unit converts the acquired first image data based on a predetermined distribution rule. A process of distributing to one of the representative representative gradation values in one color space is executed.

  The distribution is to reduce the number of gradations of the image data before and after the distribution. When the first image data is simply distributed to the neighboring representative gradation values, a smooth level of the image data before the distribution is obtained. The gradation change of the image data after the distribution with respect to the tone change is stepped. This not only deteriorates the image quality but also does not realize a unique style such as the irregular disturbance desired by the user. Therefore, the image data distribution unit causes a disturbance to the distribution rule by adding a noise component to the distribution rule. Then, the color conversion means refers to the color conversion table for the second image data corresponding to the representative gradation value in the first color space where the first image data is distributed in accordance with the distribution rule in which the disturbance occurs. And read.

  That is, according to the present invention, when the first image data distribution process is performed, a noise component is given to the distribution rule to disturb the distribution rule. Therefore, in the second image data acquired as a result of the color conversion, the phenomenon of the stepwise gradation change described above is suppressed, and the image is compared with the image represented by the first image data of the conversion source. Disturbance has occurred and a unique style with a sense of noise can be achieved. In addition, since the addition of the noise component is automatically performed together with the color conversion, it does not require a large amount of work time and proficiency from the user in order to add a noise effect to the image as in the past. .

  As a specific configuration of the present invention, the image data distribution unit includes the distribution rule that describes distribution destination data representing a representative gradation value of a distribution destination corresponding to each gradation value in the first color space. Also good. Then, at the time of the distribution, the value of the distribution destination data may be increased or decreased by adding the noise component to the distribution destination data. According to such a configuration, since the distribution destination data changes randomly according to the magnitude of the noise component, when the acquired first image data is distributed to the representative gradation value according to the distribution rule, the result of the distribution As a result, a unique style with a sense of noise is realized in the converted second image data.

  As another configuration of the present invention, the image data distribution unit acquires a printing condition for the predetermined image, generates a noise component corresponding to the acquired printing condition, and assigns it to the distribution rule. It is good. In other words, by making the size of the noise component different depending on the printing condition and providing the noise component having a size that matches the printing condition, a more satisfying style of the user is realized. The image data distribution unit can acquire the printing conditions via a UI (user interface) provided in the image processing apparatus.

More specifically, the image data distribution unit may acquire at least the size of the printing paper used for printing as the printing condition, and change the generated noise component according to the acquired size of the printing paper. Good. Further, the image data distribution means acquires part or all of various printing conditions such as the type of printing paper used for printing, the combination of ink types used for printing, the output resolution at the time of printing, and the like as the printing conditions. A noise component having a magnitude corresponding to the printing condition may be generated.
As described above, according to the present invention, the optimum noise component is selected according to the difference in specific printing conditions such as the size of the printing paper, the type of printing paper, the combination of ink types used for printing, and the output resolution. By assigning to the distribution rule, it is possible to realize an image in which irregular irregularities are appropriately generated according to the printing conditions.

So far, the invention has been described by the category of image processing apparatus. However, the contents of the image processing apparatus can be grasped as a procedure executed by each means, and can be grasped as an invention of an image processing method.
Similarly, the contents of the image processing apparatus can be grasped as an invention of an image processing program that causes a computer to execute a procedure executed by each means.

Embodiments of the present invention will be described in the following order.
(1) Outline of image processing device (2) Color correction profile (3) Contents of image processing (4) Summary

(1) Overview of Image Processing Device FIG. 1 shows a schematic configuration of a computer that functions as an image processing device according to the present invention. In FIG. 1, a computer 10 includes a CPU 11, a RAM 12, an HDD 13, a USB interface (I / F) 14, an input device interface (I / F) 15, and a video interface (I / F) 16 connected by an internal bus 10a. The HDD 13 stores various program data 13a, a plurality of image data 13b, a color conversion lookup table (LUT) 13c, a pre-conversion table 13e, and a color correction profile 13d. The color conversion LUT 13c corresponds to a color conversion table referred to in the claims.

  The CPU 11 reads out the program data 13a and executes processing based on the program data 13a while using the RAM 12 as a work area. An inkjet printer 20 is connected to the USB interface (I / F) 14, and a mouse 40 and a keyboard 50 are connected to the input device interface 15. Further, a display 60 is connected to the video interface (I / F) 16.

  FIG. 2 shows a software configuration of a program executed on the computer 10. In the figure, a printer driver P, which is one of the program data 13a, is executed on an operating system (O / S) (not shown). The printer driver P includes an image data acquisition unit P1, a color space determination unit P2, a color correction unit P3, a color conversion unit P4, a halftone processing unit P5, and a print data generation unit P6. .

  The image data acquisition unit P1 receives image data from the other application or acquires image data 13b stored in the HDD 13 in response to a print execution instruction from another application executed on the O / S. To do. In the present embodiment, the application is AdobeRGB (Adobe is a registered trademark of Adobe Systems Inc .; hereinafter, aRGB is described as an RGB color space) and RGB values in an sRGB color space (hereinafter referred to as RaGaBa, RsGsBs, respectively). Thus, it is possible to output image data in which the color of each pixel is expressed in gradation. Therefore, the HDD 13 includes image data 13b in which the color of each pixel is expressed in gradation in either the aRGB color space or the sRGB color space. Note that the image data 13b in which the color of each pixel is expressed by the RaGaBa value is simply referred to as image data 13b in the aRGB color space, and the image data 13b in which the color of each pixel is expressed by the RsGsBs value is simply in the sRGB color space. It is assumed to be image data 13b.

  The image data 13b includes a body that stores gradation data of each pixel and a header that stores incidental information of the image data 13b. The color specification indicates which color space the image data is in. Spatial information is stored in this header.

  The color space determination unit P2 analyzes the header of the image data 13b acquired by the image data acquisition unit P1, and determines whether the color space of the image data 13b is the sRGB color space or the aRGB color space. To do. For example, when the image data 13b is captured by a digital still camera that supports the aRGB color space, the color space determination unit P2 determines that the image data 13b is image data 13b in the aRGB color space. Determine.

  When the image data 13b is image data in the sRGB color space, the color correction unit P3 refers to the color correction profile 13d, corrects the image data 13b, and converts the image data 13b into image data 13b in the aRGB color space. The color correction profile 13d includes a table (color correction LUT) that defines the correspondence between gradation data before and after correction. In the present invention, in addition to the color correction LUT, the color correction profile 13d describes noise information α for determining the magnitude of the noise component N given to the pre-conversion table 13e. Details of the color correction profile 13d will be described later.

  The color conversion unit P4 targets the image data 13b in the aRGB color space and refers to the color conversion LUT 13c to convert the image data 13b into image data 13b (ink data) in the ink color space. The color conversion unit P4 performs color conversion of the image data 13b in the aRGB color space into the ink color space, so that the ink recording amount for each ink type used in the printer 20 can be specified. Thus, a print image corresponding to the image data 13b can be output. In the present embodiment, the color conversion unit P4 adds the noise component N to the pre-conversion table 13e in which the distribution rule is disturbed by adding the noise component N before the color conversion process with reference to the color conversion LUT 3c. Is used to perform pre-conversion that assigns the gradation value of each pixel of the image data 13b in the aRGB color space to another gradation value. In this sense, the color conversion unit P4 corresponds to the image data distribution unit and the color conversion unit described in the claims. The processing by the color conversion unit P4 will be described in detail later.

  The halftone processing unit P6 acquires the image data 13b of the ink color space obtained by color conversion, and performs halftone processing. In the halftone process, halftone data for specifying whether or not to eject ink droplets for each pixel and for each ink type is generated based on the image data 13b in the ink color space. For example, an error diffusion method, a dither method, or the like can be applied to the generation of halftone data. If the printer 20 can eject ink droplets of a plurality of sizes, halftone data specifying the size of the ink droplets to be ejected together with whether or not to eject ink is generated. The print data generation unit P7 rearranges the halftone data in the order of printing in the printer 20, and generates print data in a format that can be sequentially processed by the printer 20.

(2) Color Correction Profile The color correction profile 13d will be described. FIG. 3 schematically shows the data structure of the color correction profile 13d. The color correction profile 13d roughly includes a header 13d1 area for storing incidental information about the color correction profile 13d, a body area 13d2 describing the color correction LUT, and a noise information area 13d3 describing the noise information α. .

  First, the color correction LUT will be described. The color correction LUT defines a correspondence relationship between an RsGsBs value before correction and an RmGmBm value after correction (gradation data after correction with reference to the color correction LUT is expressed as RmGmBm). The color correction LUT may prescribe the corrected RmGmBm values corresponding to all combinations of RsGsBs values (256 cube combinations when RGB is expressed in 256 gradations). A corrected RmGmBm value corresponding to a typical RsGsBs value may be defined, and the corrected RmGmBm value may be calculated by interpolation calculation. The corrected RmGmBm value represents the coordinates in the aRGB color space. Therefore, the image data 13b in the sRGB color space can be converted into the image data 13b in the aRGB color space by performing correction with reference to the color correction LUT.

FIG. 4 shows the color gamut of the aRGB color space, the color gamut of the sRGB color space, and the color gamut that can be reproduced by the printer 20 in the L ^* a ^* b ^* color space (hereinafter, “ ^* ” is omitted). The planar section which cut | disconnected so that the same color gamut might be included including the L-axis and a axis | shaft (red direction) is shown. In the figure, in the high saturation region of the a-axis (red direction), the color gamut of the sRGB color space is wider on the high lightness L side than the color gamut that can be expressed by the printer 20. Further, in the design of the color space, the color gamut of the aRGB color space is wider on the high lightness L side than the sRGB color space in the high saturation region. At the maximum point Os where the saturation is highest in the sRGB color space, the difference from the color gamut of the aRGB color space is the largest, and the difference in color gamut is smaller as the saturation decreases.

  In FIG. 4, the correction amount of the RsGsBs value corrected by the color correction LUT is indicated by a vector h. The brightness h of the color near the outer edge of the color gamut of the sRGB color space is upwardly corrected by the vector h, and the RsGsBs value that has fallen in the color gamut of the sRGB color space is outside the color gamut of the sRGB color space and the aRGB table. It is shown that the RmGmBm value corresponding to the Lab value inside the color gamut of the color space is corrected. Further, the vector h has no component in the a-axis direction and is parallel to the L axis. That is, the saturation is the same between the RsGsBs value before correction and the RmGmBm value after correction, and the saturation can be maintained even if correction is performed by the color correction LUT.

  In the above description, the correction mode on the a-axis (red direction) is shown. However, in other hues, the relationship between the sRGB color space and the color gamut in which the aRGB color space can be reproduced may be different. For example, in the high saturation region of the a-axis (green direction), the color gamut of the sRGB color space is larger on the higher lightness side than the color gamut of the aRGB color space, and the color expressed by the sRGB color space is In order to correct to a color that can be expressed by the aRGB color space, the lightness L needs to be corrected downward. In this way, the color gamut of the sRGB color space and the color gamut of the aRGB color space (and the color gamut that can be reproduced by the printer 20) do not match and vary depending on the color region of interest. It is desirable to set an appropriate correction amount and correction direction according to the situation.

FIG. 5 schematically shows the flow of creating a color correction LUT.
First, a Lab value corresponding to each RsGsBs value in the sRGB color space is calculated based on the design specification of the sRGB color space. As a result, the color gamut of the sRGB color space can be specified in the Lab color space. Next, the Lab value corresponding to each RsGsBs value is corrected by giving a predetermined correction amount. In the present embodiment, the Lab value is corrected so that the Lab value corresponding to the RsGsBs value near the outer edge of the color gamut of the sRGB color space is located near the outer edge of the color gamut of the aRGB color space. That is, the RsGsBs value near the outer edge of the color gamut of the sRGB color space is corrected so as to be associated with the Lab value near the outer edge of the color gamut of the aRGB color space. In the example shown in FIG. 4, the high saturation red color is corrected near the outer edge of the color gamut of the aRGB color space by receiving the correction of the lightness L indicated by the vector h.

  Next, the color coordinates corresponding to the corrected Lab value in the aRGB color space are specified as the RmGmBm value. Since the correspondence relationship between the aRGB color space and the Lab color space can be specified by the design specification of the aRGB color space, the color coordinates corresponding to the corrected Lab value in the aRGB color space are specified as RmGmBm values. be able to. By performing the above processing, the RmGmBm value of the aRGB color space corresponding to the original RsGsBs value can be specified, and a color correction LUT can be created by describing these correspondences.

Next, the noise information α described in the noise information area 13d3 will be described.
As described above, not only the color correction LUT but also the noise information α is described in the color correction profile 13d, so that the noise component N can be added to the pre-conversion table 13e. In the present embodiment, the noise component N is generated by the following equation (1).
N = rand × α (1)
Here, rand represents a function for generating a random number. For example, if rand = [− 1, 1] is a uniform random number, numbers −1 to 1 are randomly generated. Note that a normal random number may be adopted for generation of the noise component N.

The noise information α is a coefficient multiplied by a random number. By changing this value, the magnitude of the noise component N generated as a random value can be changed. In the present embodiment, the predetermined information in the color correction profile 13d is described in association with each tag, but the noise information α is also described in correspondence with an original tag (EMSTag).
In the present embodiment, the noise information α described in the color correction profile 13d is not one, but a plurality of noise information α is described according to the difference in image printing conditions. One possible printing condition is the size of the printing paper used for image printing.

  More specifically, the noise information α is set to a larger value as the size of the printing paper is larger. In other words, if the paper size is large, the noise-like style cannot be expressed firmly in the image unless the noise component N is increased to some extent. On the contrary, if the noise component N is too large on a small paper, the image may be rough. Because it becomes too big. In the figure, different noise information corresponding to each paper size, such as noise information α1, α2,... Αn, is recorded in association with the tag EMS. For example, if the noise information for A3 size is α1 and the noise information for A4 size is α2, then α1> α2.

Note that the printing conditions considered when generating the noise component N are not limited to the size of the printing paper, but the type of printing paper, the combination of ink types used by the printer 20 (number of ink types), and the printing time. A part or all of the output resolution may be considered instead of or in addition to the size of the printing paper.
When considering the type of printing paper, a plurality of noise information α different for each type of printing paper is recorded in a specific tag of the color correction profile 13d. For example, assuming that various types of paper such as glossy paper, matte paper, and plain paper can be selected for image printing, different noise information α is set for each type of paper. Here, in consideration of paper characteristics and noise conspicuousness, the relationship of noise information α corresponding to glossy paper <noise information α corresponding to matte paper <noise information α corresponding to plain paper is established. Each noise information α is set.

  When considering the combination of ink types used by the printer 20, a plurality of noise information α different for each combination of ink types is recorded in a specific tag of the color correction profile 13d. In this embodiment, the printer 20 uses four types of ink sets of CMYK (cyan, magenta, yellow, and black). Depending on the printer model, Lc (light cyan), Lm (light magenta), and the like are added. In some cases, six types of ink sets or eight types of ink sets to which other types of inks are added may be used. In the present embodiment, basically, the larger the number of ink types to be used, the larger the noise information α is set in correspondence with the value.

When considering the output resolution at the time of printing, a plurality of noise information α different for each output image resolution is recorded in a specific tag of the color correction profile 13d. In the present embodiment, basically, as the output resolution is higher, the noise information α is recorded in association with a larger value. This is because when the output resolution is small, the noise feeling of the image is easily noticeable, whereas when the output resolution is high, the noise feeling is not noticeable.
After the color correction profile 13d having the color correction LUT and the noise information α as described above is generated in advance and stored in the HDD 13, the following image processing is executed.

(3) Flow of Image Processing FIG. 6 is a flowchart showing the content of image processing executed by the printer driver P.
First, in step S (hereinafter step description is omitted) 100, the image data acquisition unit P1 acquires the image data 13b. The image data acquisition unit P1 acquires the image data 13b of the original image and the size (output resolution) of the print image, and executes size conversion of the image data 13b according to the size conversion ratio obtained from the relative ratio thereof. May be. That is, pixel interpolation or thinning is performed as necessary.

In S110, the color space determination unit P2 analyzes the header of the image data 13b and determines whether the color space of the image data 13b is an aRGB color space or an sRGB color space.
In S120, the process is branched depending on whether the image data 13b is in the aRGB color space or the sRGB color space. If the image data 13b is image data in the sRGB color space, the process proceeds to S130.

In S130, the color correction unit P3 corrects the RsGsBs value of the image data 13b in the sRGB color space with reference to the color correction LUT described in the color correction profile 13d, and converts it into the image data 13b in the aRGB color space. To do.
In S140, the color conversion unit P4 performs color conversion on the image data 13b in the aRGB color space.

FIG. 7 is a flowchart showing details of the color conversion process in S140.
First, in S <b> 141, the color conversion unit P <b> 4 acquires image printing conditions in order to select the noise information α used to generate the noise component N. That is, the color conversion unit P4 acquires a printing condition necessary for selecting the noise information α among various printing conditions set by the user via the UI by operating the keyboard 50, the mouse 40, or the like. In the present embodiment, noise information α1 to αn that differs depending on the size of the printing paper is recorded in a specific tag of the noise information area 13d3 in the color correction profile 13d. Therefore, the color conversion unit P4 acquires the paper size set for printing by the user.

In S142, the color conversion unit P4 selects and reads out the noise information α corresponding to the acquired printing condition (printing paper size) from the noise information area 13d3 of the color correction profile 13d.
In S143, the color conversion unit P4 reads the pre-conversion table 13e from the HDD 13.

  FIG. 8 is a diagram schematically showing the pre-conversion table 13e. The pre-conversion table 13e converts input-side image data represented by n gradations into image data represented by a number of gradations smaller than n, and each gradation value on the input side and each gradation It defines a correspondence relationship (distribution rule) with data (distribution destination data) representing the gradation value of the value distribution destination in a predetermined digit part. Specifically, the pre-conversion table 13e performs gradation conversion of image data in which each component of RaGaBa is expressed in 256 gradations from 0 to 255 to image data expressed in 64 gradations from 0 to 63. The image data expressed in 64 gradations obtained by such conversion coincides with any representative coordinate in the color conversion LUT 13c in which the correspondence relationship with the ink data is recorded for each representative coordinate of RaGaBa.

  The distribution destination data defined in the pre-conversion table 13e will be described. The distribution destination data corresponding to each of the input gradation values from 0 to 255 includes a reference digit (second digit or more) indicating the gradation value of the distribution destination and a non-reference digit (not indicating the gradation value of the distribution destination). 1st digit). Input gradation values of 0 to 255 are divided into a plurality of gradation blocks for each number (four gradations) obtained by dividing the number of gradations before conversion (256) by the number of gradations after conversion (64). be able to. Therefore, in the pre-conversion table 13e, each gradation block partitioned as described above is associated with numbers 0 to 63 in order from the block on the low gradation side, and the associated numbers are assigned values in the reference digits (reference digits). Value). As a result, as shown in the figure, the reference digit value corresponding to the input gradation values 0 to 3 is 0, the reference digit value corresponding to the input gradation values 4 to 7 is 1, and the input gradation values 8 to 11 are set. The corresponding reference digit value is 2...

  However, when gradation conversion is performed using the reference digit value as it is, the conversion result is too regular, and not only the image quality by the ink data after color conversion is deteriorated but also the image desired by the user. It cannot express the style of irregular turbulence above. Therefore, in this embodiment, one digit is assigned to the lower order of the reference digit, and this is used as a non-reference digit to configure the distribution destination data. By adding a noise component N to the distribution destination data, the reference digit value can be increased or decreased. Yes.

  The method of assigning numbers to the non-reference digits of each distribution destination data is not particularly limited. As an example, in this embodiment, when the reference digit value of each distribution destination data corresponding to a certain gradation block is i, The non-reference digit values (non-reference digit values) of each distribution destination data are gradually set so that the values indicated by the respective distribution destination data in the key block fall within the range of i × 10 or more and less than (i + 1) × 10. It is specified to increase.

  In S144, the color conversion unit P4 calculates the noise component N by using the read noise information α and the above equation (1) for each distribution destination data of the pre-conversion table 13e, and the generated noise component N is calculated as the non-conversion noise component N. By assigning the value to the reference digit value, the distribution rule by the pre-conversion table 13e is changed. Each time the noise component N is calculated, it becomes a random value including positive and negative, and the value varies depending on the magnitude of the noise information α to be multiplied. Therefore, by adding the noise component N to the non-reference digit value, the upper digit of the non-reference digit, that is, the value of the reference digit can be raised or lowered.

  For example, in the case of distribution destination data “26 (reference digit value = 2, non-reference digit value = 6)” corresponding to the input gradation value “10”, if the applied noise component N is “+1”, the reference is made. The digit value does not change and remains “2”. However, if the noise component N is “−8”, the reference digit value is moved down to “1”, and the noise component N is “+8”. If so, the reference digit value is incremented by one to “3”. If the noise component N is a larger value, the reference digit value is also increased or decreased further. That is, by adding the noise component N having a random value to the distribution destination data as described above, it is possible to disturb the regularity of gradation distribution by the pre-conversion table 13e. The generation and addition of the noise component N are performed for all distribution destination data. Regardless of the addition result of the noise component N, the minimum value that can be taken by the reference digit value is 0 and the maximum value is 63.

  In the color correction profile 13d, the noise information α is recorded not according to the size of the printing paper but according to any of the printing paper type, the combination of ink types used by the printer 20, and the output resolution at the time of printing. In this case, the color conversion unit P4 acquires the type or output resolution of the printing paper among the various printing conditions set by the user via the UI in S141, or the color conversion unit P4 communicates with the printer 20 via the USB I / F 14. Information indicating a combination of ink types used by the printer 20 is acquired by bidirectional communication. In S142, the noise information α corresponding to the acquired type of printing paper, the combination of ink types used by the printer 20, or the output resolution at the time of printing is selected and read from the noise information area 13d3.

  In the color correction profile 13d, a plurality of noise information α are recorded for each printing condition such as the size of printing paper, the type of printing paper, the combination of ink types used by the printer 20, and the output resolution at the time of printing. In step S141, the color conversion unit P4 acquires each printing condition such as the size of the printing paper, the type of printing paper, the combination of ink types used by the printer 20, and the output resolution at the time of printing. In S142, the noise information α is read one by one for each printing condition from the noise information area 13d3 of the color correction profile 13d based on the acquired printing conditions. Then, one noise information is calculated by, for example, averaging the plurality of read noise information α, and a noise component N is generated based on the calculated noise information.

In S145, the color conversion unit P4 selects one pixel to be processed.
In S146, the color conversion unit P4 targets the RaGaBa values expressed in 256 gradations in each of the selected pixels, and the pre-conversion table 13e (in which the regularity of gradation distribution is disturbed) The tone conversion is performed with reference to the pre-conversion table 13e after applying noise. That is, Ra, Ga, Ba gradation values are converted into reference digit values of the distribution destination data in the pre-conversion table 13e after the addition of noise according to the values, and the converted values are converted into Ra ', Ga'. , Ba ′.

In S147, the color conversion unit P4 refers to the color conversion LUT 13c and converts the Ra′Ga′Ba ′ after the conversion into ink data that defines the ink recording amount for each ink type.
As described above, the printer 20 uses CMYK ink. For this reason, the color conversion LUT 13c defines the same color correspondence between a plurality of representative coordinates (representative gradation values) based on RaGaBa values and CMYK values (gradation values for each CYMK). Here, in the color conversion LUT 13c, 64 cubed representative coordinates specified by setting 64 grid points at approximately equal intervals on the three axes corresponding to the color components RaGaBa in the aRGB color space. Each CMYK value is recorded in association with each other. Therefore, Ra′Ga′Ba ′ after conversion obtained by referring to the pre-conversion table 13e after the noise is applied corresponds to one of the representative coordinates. The color conversion unit P4 reads out the CMYK values corresponding to the representative coordinates uniquely specified by the converted Ra′Ga′Ba ′ from the color conversion LUT 13c, and uses the read CMYK values as the pixels selected in S145. The corresponding ink data is used.

  In S148, the color conversion unit P4 selects all the pixels constituting the image data, determines whether or not the processing of S145 and subsequent steps has been executed, and if there is an unselected pixel, selects the unselected pixel. The processing from S145 is repeated, and if all pixels have been selected, the processing proceeds to 150 in FIG.

  In the above description, the color conversion LUT 13c defines the correspondence between the representative coordinates based on the RaGaBa value and the CMYK value. However, strictly speaking, the aRGB color space and the printer 20 are maintained while maintaining the substantially equal color relationship. Is defined so that color gamut adjustment with a color gamut that can be reproduced is performed. As shown in FIG. 4, there is a difference between the color gamut of the aRGB color space and the color gamut that the printer 20 can reproduce. In general, in a region where the color gamut of the aRGB color space on the input side is wider than the color gamut of the printer 20 on the output side, color suppression is performed in the same region so that it falls within the color gamut of the printer 20. Adjust the color gamut. Conversely, in a region where the color gamut of the aRGB color space on the input side is narrower than the color gamut of the printer 20 on the output side, the color gamut of the printer 20 is widely used by performing color expansion in the same region. Adjust the color gamut. That is, the color conversion LUT 13c defines a correspondence relationship between the RaGaBa value and the CMYK value with the color gamut adjustment such as color suppression and color expansion.

  In FIG. 4, the correction amount in the Lab color space corrected by the color correction LUT is represented by a vector h, and the correction amount when the color gamut is adjusted by the color conversion LUT 13c is also represented by a vector k (however, Actually, the color to be corrected by the vector k is the color in the Lab color space of the image data after the conversion by the pre-conversion table 13e). According to the vectors h and k, the Lab value corresponding to the RsGsBs value of the color located near the outer edge of the color gamut of the original sRGB color space is corrected to the vicinity of the outer edge of the aRGB color space by the color correction in S130. (Vector h), and further corrected (vector k) near the outer edge of the color gamut that can be expressed by the printer 20 by the color conversion in S140. That is, by using the color correction LUT, the Lab value corresponding to the color located near the outer edge of the color gamut of the original sRGB color space is positioned near the outer edge of the color gamut in the color space after color conversion. The color gamut that can be expressed by the printer 20 can be used effectively and widely.

  Here, if color correction is not performed in S130, color correction using the vector h is not performed, and color gamut adjustment is performed using the vector k in color conversion. Then, in FIG. 4, the red RsGsBs value that is originally near the outer edge of the color gamut of the sRGB color space is corrected to the CMYK value after being significantly corrected to the inner side of the vicinity of the outer edge of the color gamut that can be expressed by the printer 20. The same color conversion is performed. In this case, the vicinity of the outer edge of the color gamut that can be expressed by the printer 20 is not used, and the color gamut of the print image is limited.

  As described above, when the ink data after the color conversion process is obtained, the halftone processing unit P5 performs the halftone process on the ink data (S150), and the print data generation unit P6 converts the print data into print data. Processing is performed (S160). Accordingly, print data can be output to the printer 20, and the printer 20 can actually perform printing based on the print data.

  In the present invention, the RaGaBa value as the color conversion target is assigned to the representative coordinates in the vicinity of the color conversion LUT 13c, the CMYK values associated with the representative coordinates of the assignment destination are read out, and the RaGaBa value as the color conversion target is read out. CMYK values corresponding to. Therefore, the amount of calculation is small compared with the color conversion processing in which the interpolation calculation is performed with reference to the relationship between the RaGaBa-CMYK values in the representative coordinates in the vicinity of the RaGaBa value as the color conversion target, and the color conversion processing itself is speeded up. .

(4) Summary As described above, according to the present invention, each pixel of image data to be color-converted is pre-processed as a pre-stage of color conversion with reference to the color conversion LUT 13c that defines the color conversion relationship for each of a plurality of representative coordinates. When distributing to neighboring representative coordinates according to the conversion table 13e, a noise component N having a random value is assigned to the distribution destination data defined in the pre-conversion table 13e. Disturbance can be caused. Therefore, a certain degree of disturbance occurs in the result of color conversion for each pixel, and a unique style with a sense of noise desired by the user is realized in an image printed based on ink data obtained by color conversion. be able to.

  Further, according to the present invention, by performing the above-described distribution using the pre-conversion table 13e, it is possible to automatically express the noise feeling in the image. There is no requirement for the user to spend a lot of work time and proficiency to add effects. In the color correction profile 13d, noise information α for generating a noise component N having an optimum size according to various printing conditions including the size of the printing paper is recorded. For this reason, in the present invention, it is possible to automatically obtain an image that has a moderately noisy style even when the printing conditions are changed variously.

  In the color correction profile 13d, it is of great value that the color correction LUT and noise information α are recorded together. In other words, the color correction profile 13d is predetermined for a predetermined apparatus such as a computer having a color conversion LUT that defines the color conversion correspondence relationship between the pre-conversion table and the aRGB color space and the color space handled by the printer. Can be distributed through any communication network or medium. Then, in the predetermined apparatus, before the color conversion operation of the image data by the color conversion LUT, the image data can be corrected with reference to the color correction LUT, and the noise information α is read from the color correction profile 13d, It is possible to execute a noise component addition process on the conversion table.

1 is a hardware configuration diagram of an image processing apparatus according to an embodiment of the present invention. 1 is a software configuration diagram of an image processing apparatus according to an embodiment of the present invention. The figure which showed schematically the data structure of the color correction profile. The figure which shows the color gamut in the a-axis of Lab color space. FIG. 5 is a schematic diagram showing a flow of creating a color correction LUT. The flowchart which shows the content of image processing. The flowchart which showed the detail of the color conversion process. The figure which showed the pre conversion table typically.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Computer, 10a ... Internal bus, 11 ... CPU, 12 ... RAM, 13 ... HDD, 13a ... Program data, 13b ... Image data, 13c ... Color conversion LUT, 13d ... Color correction profile, 13e ... Pre-conversion table, 14 ... USB I / F, 15 ... Input device I / F, 16 ... Video I / F, 20 ... Printer, 40 ... Mouse, 50 ... Keyboard, 60 ... Display, P ... Printer driver, P1 ... Image data acquisition unit, P2 ... Color space determination unit, P3 ... color correction unit, P4 ... color conversion unit, P5 ... halftone processing unit, P6 ... print data generation unit

Claims (6)

The correspondence between the first image data in which the color of each pixel is expressed by gradation in the first color space and the second image data in which the color of each pixel is expressed by gradation in the second color space is a representative gradation value. An image processing apparatus that converts first image data into second image data with reference to a color conversion table defined for
Image data acquisition means for acquiring first image data representing a predetermined image;
When executing the process of distributing the acquired first image data to one of the neighboring representative gradation values in the first color space based on a predetermined distribution rule, a noise component is added to the distribution rule. Image data distribution means for causing disturbance in the distribution rule by,
An image processing apparatus comprising: color conversion means for reading out second image data corresponding to a representative gradation value in the first color space of the distribution destination with reference to the color conversion table.   The image data distribution means includes the distribution rule that describes distribution destination data representing a representative gradation value of a distribution destination corresponding to each gradation value in the first color space, and at the time of the distribution, the distribution destination The image processing apparatus according to claim 1, wherein the value of the distribution destination data is increased or decreased by adding the noise component to the data.   2. The image data distribution unit acquires a printing condition for the predetermined image, generates a noise component according to the acquired printing condition, and assigns the noise component to the distribution rule. The image processing apparatus according to claim 2.   The image data distribution unit acquires at least a size of a printing paper used for printing as a printing condition, and changes a noise component to be generated according to the acquired size of the printing paper. An image processing apparatus according to 1. The correspondence between the first image data in which the color of each pixel is expressed by gradation in the first color space and the second image data in which the color of each pixel is expressed by gradation in the second color space is a representative gradation value. An image processing method for converting first image data into second image data with reference to a color conversion table defined for
An image data acquisition step of acquiring first image data representing a predetermined image;
When executing the process of distributing the acquired first image data to one of the neighboring representative gradation values in the first color space based on a predetermined distribution rule, a noise component is added to the distribution rule. An image data distribution process that causes disturbance in the distribution rule,
An image processing method comprising: a color conversion step of reading second image data corresponding to a representative gradation value in the first color space of the distribution destination with reference to the color conversion table. The correspondence between the first image data in which the color of each pixel is expressed by gradation in the first color space and the second image data in which the color of each pixel is expressed by gradation in the second color space is a representative gradation value. An image processing program for causing a computer to execute a process of converting the first image data into the second image data with reference to the color conversion table defined for
An image data acquisition function for acquiring first image data representing a predetermined image;
When executing the process of distributing the acquired first image data to one of the neighboring representative gradation values in the first color space based on a predetermined distribution rule, a noise component is added to the distribution rule. An image data distribution function that causes disturbance in the distribution rules,
An image processing program for executing a color conversion function of reading second image data corresponding to a representative gradation value in the first color space of the distribution destination with reference to the color conversion table.

Images