METHOD FOR PERFORMING TRANSFORMATIONS OF COLOR DATA

Abstract

The invention includes a method for the computer-aided performance of color space transformations with high accuracy that provides a result which is as close as possible to the original. The source color space includes n colors, the target color space includes m values, and combinations of the m components of the target color space are assigned to at least some combinations of the n colors of the source color space via the transformation rule TRV.

Claims

1. Method for computer-aided execution of transformations of color data from a source color space to a target color space using a transformation rule TRV, wherein the source color space comprises n colors, which are present in combinations of color components q(1) to q(n) in each point to be represented, the target color space comprises m values that can be combined to form combinations of components z(1) to z(m), combinations of them components of the target color space are assigned to at least some combinations of the n colors of the source color space via the transformation rule TRV, characterized by the following steps: a) for a combination KB of the n colors of the source color space with color components q.sup.KB(1) to q.sup.KB (n) for which no combination of the m components z.sup.KB (1) to z.sup.KB (m) of the target color space is assigned via the TRV, select a color i of the combination KB which has a color component FA=q.sup.KB (i)>0 and for which the following conditions apply: i) the remaining combination of color components q (j not equal to i) without the component of color i is assigned a combination of the target color space with components z(1).sup.i, z(2).sup.i, . . . , z(m).sup.i, and ii) two further, with one exception mutually identical combinations of color components q(1) to q(n) exist, to each of which a combination of components z(1) to z(m) is assigned and wherein the two combinations of color components q(1).sup.1 to q(n).sup.1 and q(1).sup.2 to q(n).sup.2 differ only in that the color component q(i) of color i in the one q(i).sup.1=FA>0 and in the other q(i).sup.2=0, so that 1) the combination q(1).sup.1 to q(n).sup.1 with color component q(i).sup.1=FA is assigned a corresponding combination of the components z(1).sup.1, z(2).sup.1, . . . , z(m).sup.1 of the m values of the target color space, which forms the color data set Z1, and 2) the combination q(1).sup.2 to q(n).sup.2 with color component q(i).sup.2=0 is assigned a corresponding combination of the components z(1).sup.2, z(2).sup.2, . . . , z(m).sup.2 of the m values of the target color space, which forms the color data set Z2, iii) Calculating the ratios V(1).sup.i, V(2).sup.i, . . . , V(m).sup.i of each component z(1).sup.1, z(2).sup.1, . . . , z(m).sup.1 of the m values of the color data set Z1 to the respective component z(1).sup.2, . . . , z(2).sup.2, . . . , z(m).sup.2 of the m values of the color data set Z2, which form a set of factors V(1).sup.i=z(1).sup.1/z(1).sup.2, V(2).sup.i=z(2)/z(2).sup.12, . . . , V(m).sup.i=z(m)/z(m).sup.12, b) Applying the factors V(1).sup.i, V(2).sup.i, . . . , V(m).sup.i for transformations of combinations KB of colors of the source color space with n colors, which contain the color component FA of color i, but for which the color component q(i)=0 is set for the transformation, into the target color space with m components by multiplying the z(1).sup.i, z(2).sup.i, . . . , z(m) of the target color space resulting from the transformation by the respective factors V(1), V(2), . . . , V(m), z(m).sup.i of the target color space are multiplied by the respective factors V(1).sup.i, V(2).sup.i, . . . , V(m).sup.i.

2. The method according to claim 1, characterized in that the method is applied to partial combinations of k<n colors for a given combination of then colors with combinations of color proportions q(1) to q(n).

3. The method according to claim 2, characterized in that the method is applied to successive partial combinations with k<n colors up to k=n, the resulting factors being applied multiplicatively.

4. Method according to claim 1, characterized in that the size of the color portion of a color is taken into account for achieving a higher accuracy by preferentially selecting from the possible colors i the one with smaller color portion.

5. Method according to claim 1, characterized in that for several different possible colors i1, i2, . . . the results of the different estimates z(1).sup.i1, . . . , z(m).sup.i1 as well as z(1).sup.i2, . . . , z(m).sup.i2 etc. are averaged in a weighted manner, the weight of each estimate being selected to be greater the smaller the respective color proportion q(i1), q(i2), . . . of the color.

6. Method according to claim 1, characterized in that the components z(1) to z(m) of the target color space are device-dependent values.

7. Method according to claim 1, characterized in that the components z(1) to z(m) of the target color space are device-independent values.

8. Method according to claim 1, characterized in that the transformation rule TRV is in the form of transformation tables.

9. Method according to claim 1, characterized in that the method is carried out on a computer unit by means of control software, the computer unit comprising an input unit for providing the digital color data of the source color space of the project and an output unit for outputting the transformed values of the target color space as well as a memory on which transformation tables are stored, values for the target color space being generated for the color data of the source color space of the project by means of the control software using the transformation tables and applying the method according to claims 1 to 8 and being provided in a data set.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Further advantages and features result from the following description based on the figures. Thereby show:

[0094] FIG. 1 a flow chart explaining the process steps.

[0095] FIG. 2A flowchart for estimating multiple missing colors.

[0096] FIG. 3 the occurrence of more than 4-dimensional color data.

[0097] According to FIG. 1, there is an input combination KB of color data 101 with color components Q={q(1), q(2), . . . , q(n)} from the source color space with n colors, which is to be transformed into the target color space with combinations of m values. The non-vanishing color components of this combination form a so-called color group. In step 102, a color i is searched for from this color group, whose color component is thus FA=q(i)>0, and for which certain conditions apply: [0098] 1) for the subset Q.sup.i (103) of the color data 101, for which the color component q(i).sup.i=0 is set, an output Z.sup.i (107) from the target color space must be assigned by the transformation rule TRV 106, which thus describes the transformation result of a color group without the color i, which is to be supplemented by the influence of the color i later. [0099] 2) There must exist another input Q1 (104) for which q(i).sup.1=FA, and which is otherwise arbitrary, to which an output Z1 (108) is assigned by the transformation rule TRV 106, i.e. which describes the transformation result of this other input with the same color component FA of the color i. [0100] 3) Appropriately, an input Q2 (105) is formed that is identical to Q1 except for the portion of color i that is q(i).sup.2=0. An output Z2 (109) is assigned to this input by the transformation rule TRV 106, thus describing the transformation result of this other input without the influence of color i.

[0101] If this color i exists, the influence of its color component FA is now estimated from the ratio V.sup.i (110) of the output pair Z1 and Z2 with and without color component FA. This influence can be applied to all combinations with color share q(i)=FA for which there is an assignment without FA, in particular, of course, to the input combination KB. For this purpose, in step 111 the output Z.sup.i without the contribution of color i is multiplied component-wise by the ratio V.sup.i (110). Thus the sought result 112 is formed, which is the estimate of the target color values for the entire color group input combination KB.

[0102] In other words, the color group of the input combination is reduced in size by removing color i, under the condition that an output is assigned to this smaller color group, and then color i is added by estimation so that the original color group is complete again. According to the invention, this reduction can also be carried out several times, so that after n steps at the latest one arrives at the empty color group, the unprinted substrate, for which an assignment can be easily determined. From there, if necessary, each individual color can be added step by step until the complete color group is obtained, as shown in FIG. 2.

[0103] According to FIG. 2, there is color data 201 from the source color space with n colors, which are to be transformed and represent the input. In step 202, a color group is searched for which target color data 203 are available in the transformation rule. This color group is a subset of the color data 201 and is to be added to the full set step by step, while also adding the target color data 203 to the full output. If in step 204 the current color group is already complete, the output 205 is ready. Otherwise, one of the still missing colors i is selected in step 206. For this one, a pair of two color data sets Q1 and Q2 is now selected, namely in step 207 a color data set Q1, which contains the given color portion FA=q(i) of color i, and for which a combination of the m values z(1).sup.1, z(2).sup.1, . . . , z(m).sup.1 of the target color space, which form the color data set Z1 (208), and in step 209 the color data set Q2, which contains the same color components as Q1 for all colors except color i, while the component of color i is equal to 0, and for which a combination of the m values z(1).sup.2, z(2).sup.2, . . . , z(m).sup.2 of the target color space is present, which form the color data set Z2 (210). Both selected data sets, the color group with proportion of i and the color group without proportion of i, are color data of the n colors of the source color space. In step 211, the ratios of the m values of the color data set Z1 (108) to the m values of the color data set Z2 (210) are calculated component-wise. These ratios V.sup.i are applied component-wise to the current output in step 212 as the estimated effect of color i, and color i is added to the current color group (213). This is repeated if necessary until the color group is complete and thus the result of transformation 205 is available.

[0104] FIG. 3 illustrates the known occurrence of more than 4-color overprints, although the individual objects do not use more than 4 colors, in addition to the description on page 7. Shown is a stylized packaging design 201 with an image area 202 built up in 4 colors in CMYK and a logo area 203 with a spot color. The enlarged view 204 of the design shows that the contours of the CMYK image and spot color rectangle do not overlap, only abut. So there are no 5-color areas in the design. Only the so-called trapping 205 in print production, which is used as described to avoid register-related flashes, enlarges one of the contours and thus creates overprinting areas 206 of the spot color with CMYK, where 5 colors now appear simultaneously. Independently of this, halftoning with antialiasing 207 can also mix halftone dots from portions of the areas belonging to them, so that further 5-color pixels 208 with CMYK and spot color portions are created.

[0105] The preceding examples can be repeated with similar success by substituting the generically or specifically described reactants and/or operating conditions of this invention for those used in the preceding examples.

[0106] From the foregoing description, one skilled in the art can easily ascertain the essential characteristics of this invention and, without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions.