JP2007043613A - Color data processing method, color data processing apparatus, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of obtaining color gamut data with superior color reproducibility, by applying smoothing to the color gamut data whose color gamut is compressed and generating a color conversion table with high accuracy. <P>SOLUTION: A color data processing method is characterized, in that when a color signal in a first color gamut is converted into a color signal in a second color gamut, the first color gamut is compressed into the second color gamut as a third color gamut, and smoothing processing is applied to the third color gamut, whose color gamut is compressed to obtain a fourth color gamut. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a color conversion table creation method, a color data processing apparatus, and a program.

  In recent years, digital devices such as digital cameras and image scanners have become widespread, and digital images can be easily obtained. On the other hand, full-color hard copy technology is also rapidly developing. In particular, ink-jet printing has become widely used because the print image quality is comparable to that of silver salt photographs. In addition, networks such as the Internet are widely spread, and the environment is such that many users can connect various devices. In an environment where such input / output devices are diverse, for example, a color image signal on a monitor having a wide color gamut is often hard-copied by printers having different color reproduction capabilities. That is, color image data is frequently input / output between devices having different color gamuts.

  In such an environment, a “color management system (hereinafter also simply referred to as“ CMS ”)” is widely known when realizing the same color reproduction between different devices. FIG. 1 is a diagram showing an outline of one configuration of this CMS, and shows a CMS using a device-independent color space. For example, when an input device such as a digital camera or a scanner is connected to an output device such as a printer or a monitor, the configuration shown in FIG. 1 is considered. In this case, conversion from an input color signal to an output color signal is realized by interposing a device-independent color space according to each profile. The device-independent color space includes, for example, a CIE-XYZ color space, a CIE-L * a * b * color space, or an sRGB color space. A profile refers to a conversion formula, a color conversion matrix, or a color conversion lookup table (hereinafter referred to as a color conversion table) that takes into account the relationship between device colors and device-independent color spaces.

  A smoothing technique used when creating a color conversion table in which the relationship between the device color and the device-independent color space is described will be examined. First, details of the color conversion table will be described. A case where the device independent color space is an sRGB color space and the output device is an ink jet printer is taken as an example. In this case, the color conversion table associates colors in the printer color gamut with colors in the sRGB color space. Specifically, the color conversion table is a table in which the sRGB color space is discretely divided by grid points, and the colors of the printer color gamut are associated with the grid points. Here, in general, the shapes of the input and output color spaces often do not match. Also in the sRGB color space and the printer color gamut given as examples, the shapes are different. Therefore, when creating a color conversion table, a gamut compression technique is used to reproduce colors that cannot be reproduced between devices. Some gamut compression techniques linearly compress the input color space into the color gamut of the output device, and absorb the influence of the color gamut that differs between input and output devices. In order to associate the color in the sRGB color space after gamut compression with the color in the printer color gamut, the following method may be used. For example, when the color space of the signal received by the printer is an RGB signal, for all colors in the color gamut, the RGB signal and the color (for example, CIE-L * a * b * value) specified in the colorimetric space are determined. Corresponds in a one-to-one relationship. Therefore, the RGB color space is divided at equal intervals, for example, by 729 grid points, and the grid points are evenly distributed. Such a color patch having 729 grid points is prepared and printed by a printer. By measuring the color patch after printing, the color of the grid point represented by the printer specific RGB value (referred to as printer RGB value) is specified as, for example, the color of the CIE-L * a * b * color system I can do it. Next, the compressed grid point of the sRGB color space is converted into a color in the CIE-L * a * b * color system, and the grid point with the smallest color difference is searched from the 729 colorimetric values. By performing the interpolation calculation at the grid points around the minimum color difference point, the printer RGB values corresponding to the grid points in the sRGB color space are obtained. As described above, it is possible to create a color conversion table that describes in which color in the printer color gamut the colors in the input color space are output.

  However, various noises may be mixed in the value of the created color conversion table. Hereinafter, noise refers to a phenomenon in which smooth gradation is not obtained because tone jumps or color jumps occur because accurate grid points cannot be obtained. For example, when the input color space is gamut-compressed to the output color space, the shape of the input color space after compression may be distorted. FIG. 2 is a diagram for explaining shakiness of the input color space after compression in the L * a * coordinate system. A color gamut 201 represents a printer color gamut, and the input color space is subjected to gamut compression so as to be within the color gamut 201. A color gamut 202 represents an input color space after gamut compression. Like the area of the dotted line 203, the input color space after gamut compression does not have a smooth shape similar to the printer color gamut, and may be rattled. The cause of rattling is as follows. At the time of gamut compression, a distance T from the compression convergence point to the outermost contour of the input color space, a distance D from the outermost contour of the output color space, and a distance P to a point to be compressed (hereinafter referred to as a compression target point). is necessary. Therefore, the color gamut is analyzed by linearly connecting the adjacent outermost grid points in the input and output color spaces. FIG. 3 shows the input color space on the L * a * plane with the sRGB color space and the output color space as the color gamut of the inkjet printer. A point indicated by a white point indicates an intersection (referred to as a gamut analysis point) of a straight line connecting the hue plane of the compression target point and the outermost lattice point. The color gamut indicated by the solid line represents the sRGB color space 302 and the printer color gamut 303 with continuous gradation. For example, when the outermost points of the sRGB color space such as the points P1 and P2 are gamut-compressed, the distance from the compression convergence point O to the compression target point, the distance to the outermost contour of the printer color gamut, the sRGB color A distance to the outermost part of the space is required. When a unit vector from the convergence point toward the compression target point is defined and this vector is extended little by little, it intersects with a straight line connecting gamut analysis points. What is necessary is just to take the distance when the vector crosses this straight line. However, it is difficult to accurately obtain the distance necessary for gamut compression. Here, the distance obtained may vary depending on the amount by which the vector is extended. Furthermore, the gamut analysis points are discrete, and the color gamut 303 connecting these points with a straight line and the original smooth printer color gamut 302 have different shapes. For example, the distance D2 to the outermost contour of the printer color gamut represented by the analysis point used when compressing the point P2 may be shorter than the distance D2 'to the outermost contour of the original printer color gamut. In this case, the point after gamut compression enters inside the original printer color gamut. On the other hand, when the point P1 is gamut-compressed, the distance D1 to the outermost contour of the printer color gamut coincides with the distance from the gamut analysis point to the point O. Therefore, the point P1 is compressed to the position of the distance D1 from the compression convergence point, and the drop like the point P2 does not occur. Further, as described above, when the printer color gamut is represented by uniform 729 lattice points, the lattice point interval in the vicinity of the outermost contour is widened. For this reason, there are cases where the deviation of the distance near the outermost contour is large and the degree of rattling is also large.

  That is, since the distance necessary for gamut compression cannot be obtained accurately, the shape of the input color space after gamut compression is distorted. Furthermore, in order to know the characteristics of the printer color gamut, color measurement of the color patch is performed, and at that time, a color measurement error may be included. This color measurement error also contributes to noise.

  If a lookup table is created based on such a non-uniform color gamut, the color of the input color space and the color of the output color space are combined with the noise factor included. In other words, the finally obtained color conversion table includes the influence of noise during gamut compression. As a result, the change in the value of the color conversion table is not smooth, and a pseudo contour may occur in the image converted using this color conversion table, or the gradation may be inverted.

  Therefore, for example, Japanese Patent Application Laid-Open No. 2001-016476 discloses a method of performing smoothing using a three-dimensional filter independently of the R, G, and B channels of the color conversion table. Further, in order to prevent gray from becoming a chromatic color by the smoothing process, the region to be processed is divided into a gray region and a chromatic color region, and the relationship of R = G = B is maintained for the gray region. Smoothing is performed. In this conventional example, the smoothing process is performed after the printer RGB values corresponding to the input color space are associated. That is, smoothing is performed by taking into account a shift in searching for printer RGB values in addition to the shakiness at the time of gamut compression. For example, when the search error of the printer RGB value further increases the wobbling at the time of gamut compression, the influence of the wobbling is further increased due to both causes, so that a strong smoothing strength may be required. is there.

Further, for example, there is a method in which a non-compressed area is provided in the printer color gamut and only the input color space outside the non-compressed area is compressed with gradation in the gamut compression technique. Using this method, even if the shape and size of the printer color gamut are different, the color in the non-compressed area can be reproduced colorimetrically and the gradation can be secured. There is. When the gamut compression method as described above is used, shakiness of the shape occurs only outside the non-compressed region. Since the non-compressed area is not compressed, the colors in the printer color gamut are reproduced as they are. Here, in order to reduce shakiness outside the non-compressed region, it is effective to apply smoothing. However, in the above-described conventional example, since the smoothing is uniformly performed on the non-compressed area except for the gray area, it is impossible to perform color reproduction in which the colors in the non-compressed area are colorimetrically matched. Further, when smoothing the printer RGB value, it is complicated to determine whether or not the color is in the uncompressed region, and also includes an error when searching for the printer RGB value. Have difficulty. Furthermore, as described above, in the input color space after gamut compression, even if the area near the outermost contour is rattled, in the conventional example, whether the color of the input color space is near the outermost contour from the printer RGB value or not. It is difficult to judge.
JP 2001-016476 A

  The present invention has been made in order to solve the above-described problems, and makes it possible to obtain color gamut data with good color reproducibility because the color gamut data compressed in the color gamut is smoothed. With the goal. Another object is to make it possible to create a color conversion table with high accuracy.

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

  When the color signal in the first color gamut is converted into the color signal in the second color gamut, the first color gamut is compressed into the second color gamut to form a third color gamut. The third color gamut compressed and smoothed is subjected to a smoothing process to form a fourth color gamut.

  As described above, according to the present invention, data after color gamut compression is smoothed. As a result, it is possible to create a color conversion table that smoothes the shakiness of the input color space data after compression caused by gamut compression and realizes good gradation reproduction. In addition, by retaining the information at the time of color gamut compression, it is possible to adjust the smoothing intensity as appropriate so as to suppress the influence of noise generated according to the compression condition.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The input color space is the sRGB color space, and the output color space is the color gamut of the inkjet printer. Further, the sRGB color space and the printer color gamut are handled by a common color system. This color system is referred to as a work color system. In the present embodiment, a uniform color space (CIE-L * a * b * color system) is used. Therefore, the value of the target color space after gamut compression is a value in the CIE-L * a * b * color system. Here, the input / output color space is not limited to the above color space. The common color system is not limited to the CIE-L * a * b * color system, and can be realized if the color system is similar to the L * u * v * color system. Needless to say.

  In this embodiment, the color conversion table is a look-up table for performing color conversion processing for considering the output characteristics of the printer with respect to input sRGB values, and the sRGB color space of input data is an RGB color system. It is assumed that the data is grid point data arranged uniformly.

Embodiment 1
FIG. 4 shows the sRGB color space as a schematic diagram in the RGB color system. Here, there are nine grid points at equal intervals on the R, G, and B axes. Grid point numbers from 0 to 8 are entered. Each point of black (Bk), red (R), green (G), blue (B), magenta (M), yellow (Y), and white (W) is illustrated. Note that the cyan (C) point is directly below the white point. The axis from Bk point to B point is called Blue axis, the axis from Bk point to G point is called Green axis, and the axis from Bk point to R point is called Red axis.

  First, the description method of the color conversion table will be described with reference to FIG. In the color conversion table, the printer RGB values when the Bk point (R, G, B) = (0, 0, 0), which is the first lattice point in the sRGB color space, is first converted. To do. In FIG. 5, as an example, the printer RGB value is (0, 0, 0). Next, proceed to the point of (R, G, B) = (0, 0, 1) by increasing the lattice point number of the Blue axis by one, and increase the points of the Blue axis one by one (R, G, B) ) = (0, 0, 8). The grid point number of the Green axis is incremented by one, and the printer RGB value at the point (R, G, B) = (0, 1, 0) where the Blue axis point is returned to 0 is entered. Next, only the blue axis grid point number is advanced from 0 to 8, and then the Green axis point is increased by one. When the above is repeated and (R, G, B) = (0, 8, 8), the grid point number of the Red axis is incremented by one, and (R, G, B) = (1, 0, 0) To do. Thereafter, the grid point numbers are increased in the same order, and the corresponding printer RGB values are described in the color conversion table.

  Next, a method for acquiring control information for controlling the smoothing by gamut compressing the sRGB color space into the printer color gamut will be described with reference to FIG. First, the process proceeds to step 601 in FIG. The RGB values in the sRGB color space are converted point by point into the CIE-L * a * b * color system in the order described in the color conversion table. Thereafter, the process proceeds to step 602. In the gamut compression method described in the first embodiment, a non-compressed area for colorimetrically matching color reproduction is provided in the printer color gamut, and other colors in the input color space are transferred to a printer outside the non-compressed area. Use a method of compressing to the color gamut. Using the above gamut compression method, even if the color gamut of the output device is different, color reproduction that matches colorimetrically is achieved in the non-compressed area, and the gradation is also outside the non-compressed area. It is possible to hold. For example, if the print media is different from the color gamut shape, such as photo-only paper and matte paper, the color reproduction will be different, but the same color reproduction will be realized in the uncompressed area, and also the gradation will be retained Is possible. Furthermore, in the first embodiment, a case where smoothing is not applied to colors in the non-compressed region will be described. This is because the non-compressed area is an area provided for color reproduction that matches colorimetrically, and does not shift colors by smoothing. Note that whether or not a point in the sRGB color space is inside or outside the uncompressed region is determined by using a color gamut inside / outside determination process as described below. First, the length of a vector (referred to as a source vector) connecting a point to be determined whether it belongs to a certain color gamut and a compression convergence point set inside the color gamut is calculated. Further, the distance from the compression convergence point to the intersection of the vector (referred to as color gamut vector) from the compression convergence point to the determination target point and the color gamut surface is obtained, and the lengths of the source vector and the color gamut vector are compared. When the length of the source vector is larger than the length of the color gamut vector, it is determined that the determination target point is out of the color gamut, and when it is smaller, it is within the color gamut.

  If it is determined in step 602 that the input value is not a color in the non-compressed region, the process proceeds to step 603 and gamut compression is performed by the following method. FIG. 7 is a diagram for explaining an example of a gamut compression method for applying the present invention. Color spaces 701 and 702 are obtained by projecting the sRGB color space and the printer color gamut onto the L * a * plane. A color space 703 represents an uncompressed region in which color reproduction is performed in which the colors in the sRGB color space are colorimetrically matched. The non-compressed area has a shape similar to the printer color gamut and a size of 70%. The uncompressed area is not limited to 70% of the printer color gamut, and may be 60%, 80%, or the like. Point O represents a compression convergence point. A point 704 represents a lattice point in the sRGB color space, and a case where this point is input will be described. Since this point is a color outside the non-compressed area, gamut compression is performed in the printer color gamut outside the non-compressed area. First, the distance X between the point 704 and the point O is calculated. Further, a point 705 where the straight line connecting the point O and the point 704 intersects the outline of the sRGB color space, a point 706 where the outermost outline of the printer color gamut intersects, and a point 707 where the outline of the uncompressed area intersects are searched, and the distance is calculated. The distances from the point O are denoted as T, D, and F, respectively. Based on the relationship between the compression convergence point O and the distance, the point 704 is compressed into the printer color gamut. The point after compression is linearly compressed to a distance that can be calculated by the following compression function (1) on a straight line connecting point O and point 704.

  Here, the compression function does not need to be linear, and a multi-order function that collapses the gradation as it is located outside the color gamut, or a function similar thereto, may be used. Next, the gamut compression conditions are stored in correspondence with the compression target points. Specifically, it is determined whether or not the point to be compressed is in the non-compressed region, and the smoothing flag is described in the index format. First, in step 602, since it is determined that the point 704 is a color outside the uncompressed region, 1 is entered in the smoothing flag, and the process ends. The smoothing flag is also 1 when the input value is a point 709 representing a black point in the sRGB color space. Here, smoothing is performed when the value of the smoothing flag is 1, and smoothing is not performed when the value is 0. Next, processing when another grid point 708 is input will be described. Since the point 708 is located in the non-compressed region, gamut compression is not performed, and color reproduction that matches colorimetrically is performed. In order to clarify the explanation, the details of the gamut compression method will be described using these two points. Since the point 708 in the sRGB color space to be gamut compressed is located in the non-compressed region, gamut compression is not performed, and color reproduction that matches colorimetrically is performed. A value of 0 is entered in the smoothing flag, and the gamut compression process is completed.

  As described above, the above processing is applied to all grid points in the sRGB color space in the order described in the color conversion table to perform gamut compression, and at the same time, information used at the time of compression (hereinafter referred to as compression information). ) In the index. Further, separately from the index, the L * a * b * values of the lattice points after compression are also stored.

  Next, the smoothing method will be described with reference to FIG. In the present embodiment, the colors in the sRGB color space after gamut compression are smoothed. Since the CIE-L * a * b * color space is the work color system, the color of the sRGB color space after gamut compression is the color of the CIE-L * a * b * color system. First, the process proceeds to step 801, and the L * a * b * data in the sRGB color space after gamut compression stored in advance is advanced in the same order as when the color conversion table is created. Next, in step 802, the smoothing flag is obtained by referring to the index data corresponding to the smoothing point input in step 801 (referred to as a smoothing target point). Here, if the smoothing flag is 0, the process is terminated without smoothing. When the smoothing flag is 1, the process proceeds to step 803, where neighboring grid points used for smoothing are searched from L * a * b * data in the sRGB color space after gamut compression stored in advance. The search for nearby points is performed using the relationship in the RGB coordinate system. Let P be the position of the smoothing target point. Since the number of grid points is 9 in each of the R, G, and B coordinate systems, the grid points around the symmetry point are P + 9 × 9 (referred to as P1), P-9 × 9 (P2), P + 9 (P3), P What is necessary is just to make it the point of the position of -9 (P4), P + 1 (P5), P-1 (P6). Here, there are cases where there are no black points, white points, or points on the R, G, and B axes in the vicinity of the six points. For example, since only three points P1, P3, and P5 are obtained at black points, smoothing is performed at these three points.

Next, in step 804, a method for performing smoothing will be described. Here, a process when six points near the smoothing target point are obtained will be described. The CIE-L * a * b * values of the neighboring point of the smoothing target point, the point P _{(L P,} a _P, b P), the point P1 written as _{(L P1, a P1, b} P1). The points after P1 are also described in a similar manner. Dividing into three elements of L *, a *, and b * values, the average value at the above seven points is obtained. When the L * a * b * value of the lattice point after smoothing is (L _S , a _S , b _S ) and the smoothing intensity is written as I, it can be calculated as follows.

  Finally, the process proceeds to step 805, and the smoothing target point is replaced with the smoothed value obtained in step 804. The smoothing process as described above is performed on all the compressed sRGB lattice points. Thereafter, based on the colors in the smoothed sRGB color space, the printer RGB values are searched as described in the prior art to create a color conversion table.

  As described above, by using this embodiment, it is possible to store information as a smoothing flag for determining whether or not a compression target point is a point in an uncompressed region. Therefore, smoothing processing can be realized easily and at high speed. Further, smoothing reproduction can be realized by smoothing reproduction in the non-compressed area without smoothing, reducing the influence of shakiness caused by gamut compression in other areas. Here, in the first embodiment, the method of replacing the smoothing method with the average value has been described in order to briefly explain the principle of the embodiment. However, the present invention is not limited to this, and for example, a smoothing method may be used in which an intermediate value is replaced among the points to be smoothed and points in the vicinity thereof. Further, smoothing may be performed by applying a weight such as increasing the smoothing weight to the smoothing symmetric point and reducing the weight applied to other nearby points. Also, the neighboring points are not limited to the six points mentioned in the above example, but may be any number of lattice points.

  In addition, the present invention is not limited to the gamut compression method described in the first embodiment. For example, the “colorimetric matching gamut compression method” can be applied. In this method, when the sRGB color space is compressed into the printer color gamut, the colors in the sRGB color space in the printer color gamut are faithfully reproduced, and the other regions beyond the printer color gamut are compressed. is there. When implementing this compression method, it is necessary to determine whether or not the color in the sRGB color space is within the printer color gamut. Therefore, similarly to the case described in the first embodiment, 0 is written in the smoothing flag when the color in the sRGB color space is within the printer color gamut, and 1 is written in the smoothing flag when the color is out of the printer color gamut. Smoothing may be executed based on this flag. If the above-described smoothing is applied to the colorimetric gamut compression method, the outermost contour of the sRGB color space after gamut compression becomes smooth.

  Furthermore, an important color such as a skin color may be designed with the intention of a color after gamut compression. In that case, if a smoothing process is performed, it may shift | deviate from the intended color. Therefore, it is also effective to determine a color region of a specific important color such as a skin color and perform control not to perform the smoothing process by the smoothing flag.

  As described above, it is possible to smooth the shakiness of the input color space caused by gamut compression and create a color conversion table that maintains smooth gradation. Furthermore, since compression information can be held during gamut compression, it is possible to apply an optimum smoothing strength in accordance with the compression method. Further, by performing smoothing while maintaining lightness or saturation, it is possible to avoid the problem that the color reproduction range becomes narrow due to smoothing.

[Embodiment 2]
In the second embodiment, a method of changing the smoothing strength according to the distance between the compression target points will be described. In describing this embodiment, the gamut compression method is the same as the method described in the first embodiment.

  Here, the rattling of the input color space after the above-described gamut compression may increase depending on the distance from the compression convergence point. When analyzing the input and output color spaces, if the RGB values are divided at equal grid point intervals, the grid points outside the color gamut will be reduced when expressed in the CIE-L * a * b * color system. There is. For this reason, the interval between gamut analysis points is widened, and the error during compression is also large. Therefore, in such a case, it is effective to increase the smoothing strength according to the distance from the compression convergence point.

  With reference to FIG. 9, a method for creating an index describing the smoothing strength together with gamut compression will be described. First, in step 901, RGB values in the sRGB color space are converted point by point into the CIE-L * a * b * color system in the order described in the color conversion table, and used as input values. Next, proceeding to step 902, it is determined whether or not the input value is a color in the uncompressed region. If it is determined that the color is in the non-compressed area, the smoothing strength is set to 0 and the index is entered in step 907. This is the end of the process. Here, the smoothing method performed in the present embodiment replaces the compression target point with an average value obtained from the compression target point and neighboring grid points. Here, if the intensity is 1, the smoothing intensity is replaced with an average value, and if the intensity is 0, the grid point before the smoothing is left as it is. The smoothing strength does not exceed 1.

  Conversely, if it is determined that the color is outside the non-compressed region, the process proceeds to step 904, where the distance X between the compression convergence point and the compression target point, the distance F to the outermost contour of the non-compressed region, and the maximum in the input color space. A distance T to the outer contour and a distance D to the outermost contour of the printer color gamut are calculated. In step 905, gamut compression is executed.

  Next, in step 906, the smoothing strength as described below is described. Here, the relationship between the smoothing strength and the distance X from the compression convergence point of the smoothing target point before gamut compression shown in FIG. 10 is used. In this relationship, when the smoothing target point before gamut compression is located in the outline of the non-compressed region, that is, when the distance X is equal to or less than the distance F, it is 0, and after that, when it is located in the outermost outline of the input color space. The linear function is such that the smoothing strength is 1. Therefore, at the time of gamut compression, the smoothing strength can be calculated based on the distances of X, F, and T, and the smoothing strength is described in the index in step 907.

As described above, smoothing is performed with reference to the index describing the smoothing strength obtained during gamut compression. Here, when the smoothing intensity of the smoothing target point is 0, the next smoothing target point process is executed without smoothing. When the smoothing strength of the smoothing target point is other than 0, the following processing is performed. The CIE-L * a * b * value at the compression symmetry point P is (L _P , a _P , b _P ), and the average value P _a between the compression target point and neighboring points is (L _a , a _a , b _a ). The CIE-L * a * b * value of the point P _S after smoothing is written as (L _S , a _S , b _S ). Using the intensity I, data after smoothing can be obtained as follows. This may be replaced with the value of the grid point before smoothing.
L _S = L _P × I + L _a × (1−I)
a _S = a _P × I + a _a × (1−I)
b _S = b _P × I + b _a × (1−I)

  In the present embodiment, the smoothing strength is determined by a function as shown in FIG. 10A, but the present invention is not limited to this. For example, if the slope is smoother than the slope of the smoothing function shown in FIG. Alternatively, all the smoothing strengths outside the non-compressed region may be set to 1. In this case, a smoothing result similar to that of the first embodiment is obtained. Further, the smoothing function may be determined nonlinearly using a gamma function or the like. As described above, the smoothing strength can be controlled by the smoothing function arbitrarily defined by the user. In addition, in the present embodiment, a smoothing function having the same tendency is defined for all grid points, but an optimal function may be defined for each region. For example, when a certain hue is not smoothed, all the smoothing functions can be set to zero. Further, different smoothing functions may be set in different hue / saturation regions.

  This time, an example in which smoothing is not applied to an uncompressed region has been described. However, the present invention is not limited to this example, and the smoothing is applied to the entire region without performing inside / outside determination processing. Such a smoothing function may be applied. Further, as an improvement, it is effective to suppress the amount of smoothing because the color shift is easily perceived near the gray axis. In the smoothing function of FIG. 10C, the smoothing strength is set to 0 up to a predetermined distance G to the gray vicinity region, and thereafter, the smoothing strength is increased linearly. Thereby, the gray vicinity does not cause a color shift by smoothing. As described above, it is possible to perform a smoothing process by defining a smoothing function obtained by combining a plurality of smoothing conditions. In addition, when the user intentionally adjusts the color after compression at the time of gamut compression, it is desirable not to shift the color by smoothing. Therefore, even if it is the area | region which performed the compression, you may provide the area | region which is not smoothed.

[Embodiment 3]
In the first and second embodiments, the smoothing is performed using the average value at the smoothing target point and its neighboring points. However, if the average value is replaced with a smoothing target point, the color gamut after smoothing becomes smaller than the color gamut before smoothing. For example, if the input color space for gamut compression is a printer color gamut for photo-only paper, the color gamut is wide, so the influence of the reduction of the color gamut by smoothing is relatively small. On the other hand, in the printer color gamut for plain paper, since the color gamut is narrow, the influence of the color gamut reduction becomes large. In other words, the printer color gamut for plain paper is originally narrow, the color reproduction ability is low in both brightness and saturation, and smoothing processing narrows the area used for color reproduction in the printer color gamut. Therefore, the third embodiment proposes a smoothing method that realizes good gradation reproduction without narrowing the range used by smoothing even in a narrow color gamut.

  The smoothing method in this embodiment shall replace the average value calculated by the smoothing target point and its neighboring points, as in the above embodiment. First, a * and b * of lattice points before smoothing are held and replaced with a * and b * values of lattice points after smoothing. Thereby, smoothing in the brightness direction can be performed without changing the perceptual chromaticity. Further, smoothing is performed while maintaining lightness by setting L * to a value before smoothing. Furthermore, by maintaining the hue / saturation of the grid points before smoothing, it is possible to realize smoothing of saturation / hue maintenance. These may be determined by applying a smoothing method that maintains any value depending on the situation where smoothing is to be performed.

[Embodiment 4]
When performing the smoothing process to which the present invention is applied, the user can adjust the compression information for controlling the smoothing again. For example, a slight color shift may be easily perceived in a skin color region that is particularly important in photographic reproduction. However, if the smoothing process is applied to the skin color region, a color shift occurs. In other words, it is impossible to reproduce the color that was converted by the user during gamut compression. In order to avoid this, it is sufficient that the skin color region is not smoothed. Here, when the user defines the color gamut of the skin color area in advance, as described in the first embodiment, a series of flows in which the smoothing flag is set to 0 and the smoothing process is not performed. Can be adapted. Also in the second embodiment, the smoothing strength of the skin color area may be set to zero. On the other hand, the user may adjust how much skin color area should not be smoothed by trial and error from the printer RGB values after gamut compression. In addition, by performing a process reverse to the process of searching for the printer RGB value from the L * a * b * value, an L * a * b * value corresponding to the printer RGB value can be obtained. It is also possible to do. Here, as in the first embodiment, a method for controlling smoothing using a smoothing flag will be described as an example. Needless to say, the present invention is not limited to the case where the smoothing flag is used, and it is possible to apply a method within a range not departing from the present invention, such as using the smoothing strength as in the second embodiment.

  Assuming the above case, it is only necessary that the user can operate the compression information so that the smoothing flag of the skin color area is set to 0 after the gamut compression process is completed. Here, the gamut compression need not be repeated. This is because the user only adjusts the smoothing flag and the value obtained by the gamut compression process is the same. If gamut compression is repeated, for example, if the number of grid points in the input color space is large, it takes a lot of time for gamut compression, which is inefficient. Therefore, by separating the gamut compression process and the smoothing process, adjusting the compression information, and repeating only the smoothing process, it is possible to determine the skin color area where the smoothing is not performed. Although the skin color area has been described as an example this time, it is needless to say that the present invention is not limited to this, and an arbitrary predetermined area can be adjusted.

[Embodiment 5]
In the first, second, third, and fourth embodiments, the method of performing the smoothing process on the data after gamut compression has been described. However, smoothing can also be applied after creating a color conversion table by associating printer RGB values with data after gamut compression.

  As described in the above embodiment, at the time of gamut compression, for example, compression information such as a smoothing flag for determining whether the color is in an uncompressed region is held. Here, the data after gamut compression and the compression information have a one-to-one relationship. Furthermore, the data after gamut compression and the printer RGB values correspond one-to-one, that is, the correspondence between the compression information for controlling smoothing and the printer RGB values is self-evident. Therefore, it is possible to apply the smoothing by the compressed information to the printer RGB values. Further, by performing a process opposite to the process of searching for the printer RGB value from the L * a * b * value, a predicted L * a * b * value corresponding to the printer RGB value is obtained. Using this predicted L * a * b * value, smoothness that maintains the lightness and saturation as described in the third embodiment can be performed. In this case, it is necessary to convert the smoothed L * a * b * into printer RGB values again. Here, the smoothing flag is exemplified as the compression information, but the compression information is not limited thereto.

[Embodiment 6]
When the output device is a printer, color measurement is generally performed in order to know actual characteristics, but a measurement error at the time of color measurement may cause the noise. Further, when the color of the input color space after gamut compression is associated with the printer RGB value, an error in the calculation processing result may be included. In order to smooth these errors, the printer RGB values described in the color conversion table obtained by applying the present invention may be smoothed again by using a technique as in the conventional example. This makes it possible to create a color conversion table that can reproduce smooth gradations.

  As described above, according to the first to sixth embodiments, the input color space data after gamut compression is smoothed. As a result, it is possible to create a color conversion table that smoothes the shakiness of the input color space data after compression caused by gamut compression and realizes good gradation reproduction. In addition, by retaining the information at the time of gamut compression, it is possible to adjust the smoothing strength as appropriate so as to suppress the influence of noise generated according to the compression condition.

[Embodiment 7]
A user interface (UI) of the smoothing processing unit in the color conversion apparatus for realizing the present invention will be described with reference to FIG. In the present embodiment, as in the above-described embodiment, it is assumed that a color conversion table that links the sRGB color space and the printer color gamut is created.

  FIG. 11A shows the main dialog of the smoothing processing unit. Here, two kinds of smoothing methods, emphasizing color tone and emphasizing gradation, can be selected by radio buttons. In the method of emphasizing the color tone indicated by the radio button 1101, as described in the above embodiment, the sRGB color space data after gamut compression represented by the L * a * b * color system is smoothed. Is. When the present invention is used, for example, the non-compressed area is not smoothed, and color misregistration due to smoothing is not performed. Further, when the radio button 1102 is selected, a smoothing method emphasizing gradation can be realized. The smoothing method emphasizing gradation is to smooth the printer RGB data after creating the color conversion table. As described in the fifth embodiment, smoothing processing can be performed on printer RGB value data. According to this, it is possible to perform smoothing including noise due to errors included when searching for printer RGB values. Since the gradation property of the finally created color conversion table is improved, the gradation-oriented method is adopted. When one of two types of smoothing methods is selected in the main dialog and the execution button 1104 is pressed, a smoothing process recommended by the tool creator as optimum in each method is executed.

  Further, in the method of emphasizing color tone that reflects the characteristics of the present invention, detailed setting can be performed by pressing an option button 1103. FIG. 11B is a diagram showing a detail dialog. Here, a smoothing region, a smoothing method, and a smoothing weight are set. First, the area to be smoothed is selected from the following areas other than the colorimetric matching area (radio button 1105), the important colors other than the skin color area (radio button 1106), and the entire area (radio button 1107). Next, as an option of the smoothing method, a check button 1108 is used to select whether or not to emphasize color reproduction capability. When this check button is enabled, the radio buttons 1109 and 1110 are enabled, and the smoothing method of maintaining the brightness (radio button 1109) or saturation (radio button 1110) described in the third embodiment can be realized. Further, a smoothing weight is selected. Here, by enabling the radio button 1111, as described in the second embodiment, a method of increasing the weight of smoothing according to the distance can be selected. The radio button 1112 is selected when a uniform weight is applied to the entire area. After setting the details of the above option dialog, the set option function can be realized by returning to the main dialog and pressing the execution button 1104.

  Further, after smoothing once by a method emphasizing color tone, it is also possible to perform smoothing by a method emphasizing gradation further. Accordingly, it is possible to create a color conversion table having high gradation with a small amount of color shift.

  The smoothing method, smoothing method, and smoothing weight set in the main dialog and the option dialog set in the main dialog are not limited to this. Needless to say, it can be reflected.

[Other Embodiments]
The present invention provides a software program for realizing the functions of the embodiments in a computer in an apparatus or a system connected to the various devices so as to operate the various devices so as to realize the functions of the embodiments described above. Supply the code. And what was implemented by operating the said various devices according to the program stored in the computer (CPU or MPU) of the system or apparatus is also contained under the category of the present invention.

  In this case, the program code of the software itself realizes the functions of the above-described embodiment. The program code itself and means for supplying the program code to the computer, for example, a storage medium storing the program code constitute the present invention.

  As a storage medium for storing the program code, for example, a floppy (registered trademark) disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

  Further, by executing the program code supplied by the computer, not only the functions of the above-described embodiment are realized, but also an OS (operating system) in which the program code is operating in the computer is included in the present embodiment. It is. It goes without saying that the program code is also included in the embodiment when the functions of the above-described embodiment are realized in cooperation with other application software.

  Furthermore, the supplied program code is stored in a memory provided in a function expansion board of the computer or a function expansion unit connected to the computer. Thereafter, the CPU of the function expansion board or function storage unit performs part or all of the actual processing based on the instruction of the program code, and the functions of the above-described embodiments are realized by the processing. Needless to say, it is included in

  Although various examples and embodiments of the present invention have been described above, those skilled in the art will not be limited to the specific description and drawings in this specification, and the present invention is not limited to the spirit and scope of the present invention. It goes without saying that various modifications and changes described in the scope of the above can be made.

It is a figure which shows the outline | summary of one structure of a color management system. It is a figure for demonstrating the shape of the input color space after gamut compression. It is a figure for demonstrating the cause of the noise which arises by gamut compression. It is a figure which shows the relationship of the input color space in a RGB color system. It is a figure for demonstrating a color conversion table. 6 is a flowchart for explaining a smoothing flag in the first embodiment. It is a figure for demonstrating the gamut compression method in Embodiment 1. FIG. 5 is a flowchart for explaining a method of performing smoothing with reference to a smoothing flag in the first embodiment. 10 is a flowchart for explaining a process of creating an index describing a smoothing intensity in the second embodiment. (A) is a graph which shows the relationship between the smoothing intensity | strength and distance when smoothing is not performed in the colorimetric matching area | region in Embodiment 2. FIG. (B) is a graph which shows the relationship between the smoothing intensity | strength and distance in the case of smoothing the whole area | region of input color space in Embodiment 2. FIG. (C) is a graph showing the relationship between the smoothing intensity and the distance when smoothing is not performed in the vicinity of the gray axis in the second embodiment. (A) is a figure showing the user interface of the profile creation tool carrying this invention. The main dialog is shown. (B) is a figure showing the user interface of the profile creation tool carrying this invention. Shows the options dialog.

Claims (55)

  When the color signal in the first color gamut is converted into the color signal in the second color gamut, the first color gamut is compressed into the second color gamut and the color gamut is converted into the third color gamut. A color data processing method comprising: compressing and performing a smoothing process on the third color gamut subjected to the color gamut compression to obtain a fourth color gamut.   When acquiring a grid point that divides the first color gamut with discrete grid points, and converting the color signal of the grid point into a color signal in the second color gamut, The grid point is compressed to the second color gamut to be the third color gamut grid point, and the color gamut compressed grid point is smoothed to be the fourth color gamut grid point. Characteristic color data processing method.   The smoothing holds compression information when the data of the first color gamut is compressed into data of the second color gamut, and performs the smoothing based on the compression information. The color data processing method according to claim 1.   4. The color data processing method according to claim 1, wherein the compression information is at least one of a non-compressed area, a gray axis vicinity, a flesh color area, and an important color predetermined area.   5. The color data processing method according to claim 1, wherein the compression information is a variable for determining whether or not the color is in a region not subjected to smoothing.   6. The color data processing method according to claim 1, wherein the compression information is a smoothing strength.   The smoothing strength is a distance from a compression convergence point to a point of the first color gamut to be compressed, or an outermost contour of the first color gamut on a straight line connecting the compression convergence point and the point to be compressed. Or the distance to the outermost contour of the second color gamut on the straight line connecting the compression convergence point and the grid point to be compressed, or the outermost contour of the region that reproduces the colorimetrically consistent color 7. The color data processing method according to claim 6, wherein the color data processing method is based on any one of the distances.   The smoothing intensity increases as the distance from the compression convergence point to the point of the first color gamut to be compressed or the distance from the compression convergence point to the point of the second color gamut to be compressed increases. 8. The color data processing method according to claim 6, wherein the color data processing method is performed.   9. The color data processing method according to claim 6, wherein the smoothing intensity can be a smoothing intensity function defined in advance as a function of the distance.   The color data processing method according to claim 9, wherein a plurality of smoothing intensity functions are held according to regions.   11. The color data processing method according to claim 1, wherein the smoothing is performed by selecting a method of replacing with an average value at a point near the smoothing target point, or replacing with a center value of a nearby point. .   12. The color data processing method according to claim 1, wherein the smoothing applies the smoothing to at least one element of a color signal constituting a color gamut.   13. The color data processing method according to claim 1, wherein the smoothing maintains at least one of saturation, brightness, hue, or perceived brightness.   14. The color data processing method according to claim 1, wherein the smoothing process is executed again on the conversion table for converting the data of the first color gamut into the data of the second color gamut.   The smoothing reprocessing is performed on either the color signal converted into the data of the second color gamut converted by the conversion table or the color signal of the color gamut represented by the lightness and the perceptual chromaticity. 15. The color data processing method according to claim 1, further comprising:   In the color gamut compression, when the data of the first color gamut is converted into the data of the second color gamut, the data of the first color gamut is converted to a predetermined area in the second color gamut, 16. The color reproduction approximated by colorimetry is performed, and a color outside the predetermined range of the first color gamut is compressed into data of the second color gamut outside the predetermined range. Color data processing method.   17. The color data processing method according to claim 1, wherein a smoothing method is selected from a user interface.   18. The color data processing method according to claim 17, wherein the smoothing method selection selects a smoothing region, a smoothing method, and a smoothing weight.   When the color signal in the first color gamut is converted into the color signal in the second color gamut, the first color gamut is color gamut compressed to the second color gamut to obtain the third color gamut. The color obtained by acquiring the compression information in the color gamut compression, processing the compression information, and smoothing the third color gamut based on the processing information to obtain the fourth color gamut Data processing method.   The first color gamut is divided by discrete grid points, and when the color signal of the grid point is converted into the color signal in the second color gamut, the grid point of the first color gamut is converted to the second color gamut. To a grid point of the third color gamut, smoothing the grid point compressed by the compression of the color gamut to obtain a grid point of the fourth color gamut. The color data processing method according to claim 19.   In the color gamut compression in which the first color gamut is compressed into the second color gamut, the smoothing holds compression information, processes compression information to process the compression information, and processes the processed information. A color data processing method characterized by performing smoothing.   22. The color data processing method according to claim 19, wherein the compression information is at least one of an uncompressed area, a gray axis vicinity, a skin color area, and a predetermined area of important colors.   23. The color data processing method according to claim 19, wherein the compression information is a variable for determining whether or not the color is in a region not subjected to smoothing.   24. The color data processing method according to claim 19, wherein the compression information is a smoothing strength.   The smoothing strength is a distance from a compression convergence point to a point of the first color gamut to be compressed, or an outermost contour of the first color gamut on a straight line connecting the compression convergence point and the point to be compressed. Or the distance to the outermost contour of the second color gamut on the straight line connecting the compression convergence point and the grid point to be compressed, or the outermost contour of the region that reproduces the colorimetrically consistent color The color data processing method according to claim 24, wherein the color data processing method is based on any one of the distances.   The smoothing intensity increases as the distance from the compression convergence point to the point of the first color gamut to be compressed or the distance from the compression convergence point to the point of the second color gamut to be compressed increases. 26. The color data processing method according to claim 24, wherein the color data processing method is performed.   27. The color data processing method according to claim 24, wherein the smoothing intensity can be a smoothing intensity function defined in advance as a function of the distance.   28. The color data processing method according to claim 24, wherein a plurality of the smoothing intensity functions are held depending on a region.   29. The color data processing method according to claim 24, wherein the smoothing is performed by selecting a method of replacing with an average value at a point near the smoothing target point, or replacing with a center value of a nearby point. .   30. The color data processing method according to claim 19, wherein the smoothing applies the smoothing to at least one element of a color signal constituting a color gamut.   31. The color data processing method according to claim 19, wherein the smoothing maintains at least one of saturation, brightness, hue, or perceived brightness.   32. The color data processing method according to claim 19, wherein the smoothing process is executed again on the conversion table for converting the data of the first color gamut into the data of the second color gamut.   The smoothing reprocessing is performed on either the color signal converted into the data of the second color gamut converted by the conversion table or the color signal of the color gamut represented by the lightness and the perceptual chromaticity. 33. The color data processing method according to claim 19, wherein the color data processing method is performed.   In the color gamut compression, when the data of the first color gamut is converted into the data of the second color gamut, the data of the first color gamut is converted to a predetermined area in the second color gamut, 34. The color reproduction approximated by colorimetry is performed, and the color outside the predetermined range of the first color gamut is compressed into the data of the second color gamut outside the predetermined range. Color data processing method.   35. The color data processing method according to claim 19, wherein a smoothing method is selected from a user interface.   36. The color data processing method according to claim 19, wherein in the smoothing method selection, a smoothing region, a smoothing method, and a smoothing weight are selected.   When converting the color signal in the first color gamut into the color signal in the second color gamut, the first color gamut is compressed to the second color gamut and the color gamut is compressed to the third color gamut. And obtaining color conversion data obtained by converting the third color gamut into a color of the second color gamut, and smoothing the obtained color conversion data. Obtain data of grid points that divide the first color gamut with discrete grid points,
Compressing the grid point data of the first color gamut into grid point data of the second color gamut;
Using the grid point data, create a conversion table that converts the color signal of the first color gamut into color data of the second color gamut,
A color data processing method for smoothing color conversion data of the color conversion table.   In the color gamut compression in which the first color gamut is compressed into the second color gamut, the smoothing holds compression information and performs smoothing based on the held compression information. The color data processing method according to claim 37 to 38.   40. The color data processing method according to claim 37, wherein the compression information is at least one of a non-compressed area, a gray axis vicinity, a skin color area, and a predetermined area of an important color.   41. The color data processing method according to claim 37, wherein the compression information is a variable for determining whether or not the color is in a region not subjected to smoothing.   42. The color data processing method according to claim 37, wherein the compression information is a smoothing strength.   The smoothing strength is a distance from a compression convergence point to a point of the first color gamut to be compressed, or an outermost contour of the first color gamut on a straight line connecting the compression convergence point and the point to be compressed. Or the distance to the outermost contour of the second color gamut on the straight line connecting the compression convergence point and the grid point to be compressed, or the outermost contour of the region that reproduces the colorimetrically consistent color 43. The color data processing method according to claim 42, wherein the color data processing method is based on any one of the distances.   The smoothing intensity increases as the distance from the compression convergence point to the point of the first color gamut to be compressed or the distance from the compression convergence point to the point of the second color gamut to be compressed increases. 44. The color data processing method according to claim 42, wherein the color data processing method is performed.   45. The color data processing method according to claim 42, wherein the smoothing intensity can be a smoothing intensity function defined in advance as a function of the distance.   46. The color data processing method according to claim 42, wherein a plurality of the smoothing intensity functions are held depending on a region.   47. The color data processing method according to claim 37, wherein the smoothing is performed by selecting a method of replacing with an average value at a point near the smoothing target point or replacing with a center value of a nearby point. .   48. The color data processing method according to claim 37, wherein the smoothing applies the smoothing to at least one element of color signals constituting a color gamut.   49. The color data processing method according to claim 37, wherein the smoothing maintains at least one of saturation, brightness, hue, or perceived brightness.   50. The color data processing method according to claim 37, wherein a smoothing method is selected from a user interface.   51. The color data processing method according to claim 37, wherein the smoothing method selection selects a smoothing region, a smoothing method, and a smoothing weight.   When the color signal in the first color gamut is converted into the color signal in the second color gamut, the first color gamut is compressed into the second color gamut to form a third color gamut. A color data processing apparatus comprising: a color gamut compressing unit that compresses; and a smoothing unit that performs a smoothing process on the third color gamut that has undergone the color gamut compression to obtain a fourth color gamut.   A computer-readable program for executing color data processing according to claim 1. Compress data in the first color gamut into data in the second color gamut,
A color data processing method, wherein smoothing processing is performed on the compressed data of the second color gamut to obtain data of a third color gamut. Obtain data of grid points that divide the first color gamut with discrete grid points,
Compressing the grid point data of the first color gamut into grid point data of the second color gamut;
A color data processing method, wherein smoothing processing is performed on the compressed data of the lattice points of the second color gamut to obtain third color gamut data.

Images