Black separation strategies in colour reproduction.
The main reason for adding black ink in four-colour CMYK reproduction process is because of the limited maximum density achieved with only CMY colours (Mortimer, 1998). Besides enhancing the density range, addition of black ink also improves the contrast and reproduction of shadow details. As the speed of printing machines grew higher, the total coverage amount was unable to dry in necessary time, what caused various technical and procedure difficulties. With a technique known as under colour removal (UCR), it become possible to replace some proportions of the three process colour inks, in neutral and near neutral areas of an image, with black ink. Implementation of UCR principles, solved numbered problems, including drying difficulties associated with piling four colours on a sheet while all are wet, and make the control of the printing process less difficult (Kipphan, 2001). Other advantages of UCR can include sharper rendering of fine detail, better consistency in the grey scale and savings in ink costs. Since the impressions of sharpness and resolving power are depended almost entirely on differences in luminance, rather than colour, the variations in luminance in reproduction are controlled in large measure by the black separation (Yule, 2000). That is why it is possible, by applying UCR, to obtain sharper rendering of fine detail in image. Also, UCR makes colour balance in neutrals more stable during printing, since using less of the process colours minimizes the hue shifts that can appear if they fluctuate (Green, 1999). Grey component replacement (GCR) is a further implementation of the UCR principle, reducing the grey component from all colours in a reproduction (not just neutrals), and replacing them with black ink (Mortimer, 1998).
UCR was still possible to be used in photomechanical operations as a separate mask made, but also was implemented as a hardware function in older analogue scanners. It was practised for grey scale or neutral tones defined as C M Y coverage combination to be substituted with black (K) coverage in desired amount (Yule, 2000). This is quite important as reproductions are often judged on their neutral or grey regions. Digital scanners were able to distinguish reduction for grey scale (UCR) and other tertiary colours (GCR), also applied in various graphic arts computer programs. Contemporary approach primary considers that as an achromatic substitution process for both principles.
Theoretically it is a simple process of substitution, but due to non ideal inks and substrates, various physical and optical deviations such as additivity failure, light scatter, boundary effects, have influence on the final result. One combination corresponds only to the target printing profile. In various manuals for processing in graphic arts (scanners or operating programs) the achromatic possibility is offered, but the amount of substitution is gave up to the technologist, or eventually some value for concerned process is recommended. In today's practice when ISO standardisation (ISO 12647-2, 1996) recommends achromatic methods, it is of significant interest to get more information about these issues.
In the experimental part of this work different strategies for generation of black separation were used and tested. For expressing colour and for the purpose of objective colour measurements the CIELAB colour space was used. This is currently the most important colour space based on the opponent-colour theory (Berns, 2000). The [L.sup.*], [a.sup.*] and [b.sup.*] coordinates can be calculated from the tristimulus values X, Yand Z normalized to the white by equations (1)-(3):
[L.sup.*] = 116[(Y/[Y.sub.n]).sup.1/3] -16 (1)
[a.sup.*] = 500[(X/[X.sub.n]).sup.1/3]--[(Y/[Y.sub.n]).sup.1/3]] (2)
[b.sup.*] = 200[(Y/[Y.sub.n]).sup.1/3]--[(Z/[Z.sub.n]).sup.1/3]] (3)
where [X.sub.n], [Y.sub.n] and [Z.sub.n] are tristimulus values of light source. Other transformation used in the experiment was; linear transformation of XYZ to ISO-RGB space (Sharma, 2003), given by matrix (4), for D65 illuminant,
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (4)
RGB to CMY transformation, given by equations (5) and CMY to CMYK transformation, given by equations (6)
C = 1- (R/255) M = 1- (G/255) Y = 1- (B/255) RGB values from 0 to 255 CMY results from 0 to 1 (5)
C = (C-K)/(1-K) M = (M-K)/(1-K) Y = (Y-K)/(1-K) K = min (C,M,Y) CMYK and CMY values from 0 to 1. (6)
Two colour patches, brownish (in text: Colour 1) and greenish (in text: Colour 2), were programmed on monitor for this experiment. Two different strategies for generation of black separation were used, with various combinations of achromatic substitutions. The first combination was standard substitution with black (UCR 50%). In the second combination, the output device profile values were used for calculating the amount of the achromatic substitution. All colour patches were printed on Xeikon 32D digital printer.
The workflow of the experiment is shown in Figure 1.
[FIGURE 1 OMITTED]
So, both programmed colour patches were reproduced in three different ways: only by three process colours (in text: CMY), and by four process colours with different amount of the achromatic substitutions, as explained earlier (in text: CMYK standard, and CMYK by profile). The CIE [L.sup.*][a.sup.*][b.sup.*] values of programmed colour patches are shown in Table 1.
Printed colour patches were measured by spectrophotometer (X-Rite, Eye-one Pro) and spectroradiometer (Ocean Optics, Type S2000) which measure differences in response in the area from visible to 1000 nm. It was interesting to see what is happening with process colours in IR spectrum (from cca. 750 to 1000 nm).
The CMYK coverage of reproduced colour patches is shown in Tab.2. In Tab.3. the calculated colour differences (AE*94), and differences in lightness ([DELTA][L.sup.*]) and chroma ([DELTA][C.sup.*]) between CMY and CMYK reproduced colour patches are shown.
The reproduced colour patches (screen ruling: 60 lpcm, elliptical) were viewed by microscope to visually analyze the black coverage. In Fig.2. the microscopic photographs of three printed patches of Colour 1 was shown as an example.
[FIGURE 2 OMITTED]
In Fig.3. and Fig.4. the relative response of reproduced colours is shown.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
In Fig.3. it is interesting to notice that all process colours, except pure black (K), used in the experiment, have some response in region from 700 to 1000 nm (IR region). Two black patches--one printed with 3 process colours (CMY) and other printed with pure black ink (K) visually look similar, but have different response in IR part of the spectrum. This characteristic could be used when planning the black separation strategy for colour reproduction of security graphics (Ziljak at al., 2008).
4. Results and discussion
From the obtained results, it is obvious that achromatic model of reproducing has various interesting features. In UCR reproduction process the total coverage amount is reduced and black coverage increases. ISO standardization for chromatic reduction means coverage decrease. This process at the same time lowers technical and technological requirements, where colorimetric deviations become more stable. The press-related colour variations and colour shifts related to metamerism (illuminant or observer) are minimized when achromatic component substitution techniques are used. If the black is the last ink in the printing sequence, colour balance shifts caused by typically imperfect transparency inks (especially yellow) will also be minimized.
But in practice, it is known that relatively high achromatic substitution can cause certain deviations and unwanted effects. Higher black ink substitution can lead to some higher reflection in dark areas, and a possible reduction of contrast, as well as cromaticy decrease. Of course, the main assumption is that reproduced colours with acromatic principles implementation will not vary from the ones realized without reduction. Applying about 50% achromatic component reduction has been found to give optimum results in most images. In certain situations the theoretical black amount reduction can be even exceeded. Some materials and equipment providers suggest some figures, but for each reproduction system it has to be inspected separately. The explanation for various deviations is in optical and physical properties of dyes-colorants and substrates involvement, unwanted light absorption, additivity failure and various boundary effects.
It is evident that process inks in observed reproducing system in visible part of the spectrum have common reflections, but in IR part black ink expresses much lower reflection (higher absorption). In combinations where chromatic part of selected (tertiary) colour patch was partially changed with black in the means of implemented acromatic reduction, in visible part the reproduction is similar, meaning acceptable colour differences between patches. In the extended part of the spectrum up to 1000 nm the combinations containing black coverage can be distinguished, in a way where the higher black coverage amount is applied, the lower IR response of the patch is achieved.
Achromatic reduction as a part of graphic reproduction is a process can pass various benefits. According to the specific characteristics of the output system and the calculation procedure used, acceptable chromatic yield can be substituted with black without chromaticy deviations or quality losses. The area of acceptable combinations of higher chromatic substitutions has to be specified experimentally.
The behaviour of the reproduction process inks in extended spectral regions, e.g. IR part of the spectrum, has to be defined separately. In the observed system absorption of black coverage in the combination is significant, and can provide distinction in IR part of combined visually similar coloured patches. That phenomenon could be interested in various technical applications, including security of valuables.
The special contribution of this research is in the experimental measurements of prints, made with spectroradiometer in region from 400 to 1000 nm wavelength. Those measurements have shown some specific behaviour of process inks in IR part of the spectrum, which could not be foreseen by classical methods. In this way the black separation strategies for colour reproduction could be improved and even used in some specific fields of graphic reproduction, like security graphics.
It should be noted that all the results gained in experimental part of this work are closely connected with the printing system used. If the device, inks, substrates or other conditions change, the black separation strategy must be modified and specified for specific printing process and conditions.
Further research will lead to determination of black separation by various device profiles. Also, future research will include possible metameric effects at tertiary colours as well it's application in graphic reproduction.
Berns, R.S. (2000). Billmeyer and Saltzman's Principles of Color Technology, John Wiely & Sons Inc., ISBN 0-471-19459-X, USA
Green, P. (1999). Understanding Digital Color, Pira International, ISBN 1-85802-450-1, Surrey, UK
ISO 12647-2 (1996). Graphic technology-Process control for the manufacture of half-tone
Kipphan, H. (2001). Handbook of Print Media, Springer, ISBN 3-540-67326-1, Heidelberg, Njemacka
Mortimer, A. (1998). Colour Reproduction in a Digital Age, Pira International, ISBN 1-85802-217-7, Surrey, UK
Yule, J.A.C. (2000). Principles of Color Reproduction, GATF Press, Pittsburgh, USA
Ziljak, I.; Pap, K.; Ziljak Vujic, J.; Bogovic, T. & Plehati, S. (2008). Pseudo color in infrared design, Proceedings of the 10th International Design Conference, Ziljak, V. (Ed.), pp.1497-1501, ISBN 978-953-96020-8-4 (Volume 3), Cavtat, May 2008, Faculty of Graphic Arts, Zagreb
This Publication has to be referred as: Agic, D[arko]; Strgar Kurecic, M[aja]; Mandic, L[idija] & Pap, K[laudio] (2009). Black Separation Strategies in Colour Reproduction, Chapter 01 in DAAAM International Scientific Book 2009, pp. 001-008, B. Katalinic (Ed.), Published by DAAAM International, ISBN 978-3-901509-69-8, ISSN 1726-9687, Vienna, Austria
Authors' data: Univ.Prof. Agic, D[arko]; Dr. Sc. Strgar Kurecic, M[aja]; Dr. Sc. Mandic, L[idija]; Dr. Sc. Pap, K[laudio], University of Zagreb Faculty of Graphic Arts, Getaldiceva 2, 10000 Zagreb, Croatia, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
Tab. 1. CIE L* a * b * values of programmed colour patches CIELAB Colour 1 Colour 2 L * 45 45 a * -5 -32 b * 27 29 Tab. 2. CMYK coverage of reproduced colour patches Reproduction Colour 1 Colour 2 strategy C M Y K C M Y K CMY 64 66 80 0 77 64 81 0 CMYK standard 32 37 76 32 68 32 79 32 CMYK by profile 56 46 82 40 62 52 69 44 Tab. 3. Calculated colour differences ([[DELTA]E.sup.*.sub.94]), and differences in lightness ([[DELTA]L.sup.*]) and chroma ([[DELTA]C.sup.*]) between CMY and CMYK reproduced colour patches Colour difference Colour 1 between CMY and CMYK patches [[DELTA]L.sup.*] [[DELTA]C.sup.*] CMYK standard 3.41 5.55 CMYK by profile 4.19 8.40 Colour difference Colour 1 Colour 2 between CMY and CMYK patches [[DELTA]E.sup.*.sub.94] [[DELTA]L.sup.*] CMYK standard 4.91 0.72 CMYK by profile 7.23 2.98 Colour difference Colour 2 between CMY and CMYK patches [[DELTA]C.sup.*] [[DELTA]E.sup.*.sub.94] CMYK standard 2.78 2.08 CMYK by profile 7.29 4.04
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Chapter 1|
|Author:||Agic, D.; Strgar Kurecic, M.; Mandic, L.; Pap, K.|
|Publication:||DAAAM International Scientific Book|
|Date:||Jan 1, 2009|
|Next Article:||The software application for mathematic description of extraction process course.|