GB2240897A - Color correction and printing system for reproduction of computer-generated images - Google Patents

Abstract

A hue value is developed from R, G and B intensity values of an individual pixel of an original image such as produced on the screen of a computer monitor and is used as a control in conjunction with independently developed tone, intensity and cleanness control values and terms in producing output C, M Y chromatic values, a separate black output value being separately developed from the R, G and B values. Such control values and terms are developed according to predetermined functions, through modifiable lookup tables or the equivalent. to allow use of calibration procedures which converge rapidly to an accurate color reproduction. <IMAGE>

Description

2:2-ú10ac-3'7 A_ 1 COLOR PRINTING SYSTEM USABLE FOR REPRODUCTION OF

COMPUTER-GENERATED IMAGES

BACKGROUND OF THE INVENTION

1. Field of-the Invention

This invention relates to a color printing system and more particularly to a system which is usable to obtain improved accuracy and reliability in the reproduction of images generated by computers and displayed on color monitors. The system of this invention minimizes the time required for processing operations, minimizes memory requirements and otherwise facilitates the case and accuracy with which prints can be produced having a visual match to original images produced on a color monitor screen. The system also provides substantial improvements with respect to simplicity of operation and ease of use, particularly with respect to providing calibration procedures which permit direct and rapid convergence to an accurate calibration.

is 1 2. Background of the Prior Art

As disclosed in my aforesaid prior application, various methods had theretofore been used or proposed for reproducing images of a type which are generated by an artist on a color monitor screen of a computer using "paint" programs and the like. One method which had been used involved the sending of a digital image to a color film recorder to produce a transparency which was then placed on a conventional color separation scanner for processing of color separation half-tone films for making lithographic printing plates. That type of operation was cumbersome and more importantly, did not reliably produce an accurate color reproduction.

There had also been a number of proposals for converting red, green and blue or 11RGBn values to cyan, magenta, yellow and black or 1ICNYM values, ostensibly appropriate for use with a computer system. However# the conversion methods as proposed in the prior art were quite complicated and cumbersome and were such that it would appear to be very difficult to obtain reliable results in attempting to convert computer-generated or similar signals to signals appropriate for color print reproductions.

In my aforesaid prior application, a system is disclosed for quick and accurate conversion of data such as supplied to a computer monitor screen into data for control of printing operations, to obtain accurate reproduction of an original image. In that system, chromatic output values are derived from the values of the component colors of a pixel of an original image such as__ produced on a screen of a color monitor. Tone correction values, preferably determined through look up tables, are used in producing the chromatic output values and are such as to provide a gray balance throughout a full range of 1 4 19 grays. With the system, an accurate gray balance is obtainable from color inks, without a black ink, and through a range which is quite wide, but black may be added to provide optimum contrast and otherwise facilitate an accurate reproduction or the original image. As disclosed, black is added as a function of a gray factor corresponding to the screen color which has the lowest effective intensity value.

As also disclosed in my aforesaid prior application, color correction systems are usable after taking steps to obtain an accurate gray balance, to provide accurate reproduction of colors of an original image, using color correction values which include scale values which operate in a linear fashion to obtain a balance which is close to the optimum balance. In addition, the color correction values include selective values which are added to obtain the optimum balance and highly accurate reproduction of colors. The selective values operate with respect to the screen and ink colors and are also preferably operative with respect to selected other colors such as light red, purple and brown.

Important addtional features of the system of my prior application relate to the use of a view box for viewing a reproduction in side-by-side relation to a monitor screen under standard lighting conditions and to calibration procedures which are readily followed and which provide for a rapid convergence to an accurate calibration of the system and highly reliable reproductions of original images.

SUMMARY OF THE INVENTION

This invention was evolved with the general object of improving upon the system of my aforesaid prior application.

A In a system provided in accordance with the invention, a number of the advantageous features of the prior system are retained but changes are made which greatly simplify operational and calibration procedures so as to much more readily obtain an accurate color correction. A most important change relates to the development of a hue value and the determination of output color data as a function thereof. In particular, a hue value is determined which lies in a range of from zero to 360 degrees and which corresponds to an angular rotation from a certain reference hue such as, for example, a pure red hue. Thus, for example, a value of zero or 360 degrees may indicate red, a value of 120 degrees may indicate green and a value of 240 degrees may indicate blue.

Preferably, and in accordance with a specific feature of the invention, a hue term is developed for each of three primary output colors used in reproduction, as a predetermined function of the hue value. Each such hue term may be produced, for example, using a look-up table and it is then used to control the corresponding output color.

For example, the system of this invention may be used to for control of application of cyan, magenta and yellow or OCHYn ink colors and, like that of my aforesaid prior application, may advantageously be based in part upon an assumed approximate relationship of CRY ink colors to the RGB screen colors. The assumed relationship is that cyan is the negative or inverse of red, that magenta is the negative or inverse of green and that yellow is the negative or inverse of blue. In determining each of the CMY values, one of the RGB values has an inverse relation thereto and is used as a primary value. In particular, the R, G and B values are used as primary values for 4 1 determination of the C, M and Y output values, respectively, and facilitate determination of values to obtain an approximate color match. In the system of this invention, the hue terms are used to modify or correct terms obtained from the RGB values, to control the Cj M and Y output values.

Further features relate to the determination and use of cleanness and intensity terms in conjunction with the hue terms to develop colorcorrected output data. For example, the cleanness and intensity terms may preferably be used for modification of terms obtained from the RGB values, to control the C, M and Y values.

These and other objects, features and advantages of the invention will become more fully apparent from the following detailed description taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic block diagram of a color print reproduction system usable in the practice of the invention; FIG. 2 is a flow diagram illustrating processing operations performed in accordance with the invention; FIG. 3 is a graph illustrating the relationship of a term used for printing black to a gray factor; FIG. 4 is a graph illustrating the relationship of a color removal term to a gray factor; FIG. 5 is a graph illustrating the modification of certain terms for tone reproduction in three colors; 1 FIG. 6 is a graph illustrating a further modification of terms for obtaining a gray balance; and FIG. 7 is a sample diagram indicating the type of relationship obtained between a correction coefficient and a hue value with the system of the invention.

DESCRIPTION OF A PREFERRED EMBODIMENT

In Figure 1, reference numeral 10 generally designates a system usable in the practice of the invention. An original image of a desired form is produced on the screen of a color monitor 11 using a computer 12, in a arrangement such as shown in my aforesaid prior application, the disclosure of which is incorporated by reference. Data as each pixel of the image, in the form of a group of red, green and blue pixel data, are then stored on a suitable storage medium, such as a magnetic disc or tape, using a storage unit 13. As indicated by dotted line 14, the disc, tape or other storage medium may then be physically transferred to a storage unit 15 which is then operated to transfer the data from the storage medium to the memory of a second computer 16. Computer 16 processes the red, green and blue pixel data and develops output data for use in reproduction of the image, the computer 16 being supplied with calibration data which define the relationship of characteristics of the monitor 11 and known or assumed characteristics of inks to be used in a half-tone reproduction of the original image.

The output data developed by the computer 16 includes data defining the black component and at least three color components for control of halftone reproduction of the image. For example, the output data may control cyan. magenta and yellow inks, in addition to black ink. However, the invention is also usable for control of printing processes which use additionalInks such as light cyan and pink inks, for example, and is not limited to control of four ink processes.

The output data produced by the computer 16 may be stored by a storage unit 17 on a suitable medium which, as indicated by dotted line 18, may then physically transferred to another storage unit 19. Storage unit 19 is operative to read out the output data a standard type of color processing system 20 which is coupled to a halftone film recorder 21. Films produced by the recorder 21 are then used to produce one or more reproductions, using a press 22 which may be either a production press or a proof press.

As indicated by dotted line 23, a reproduction produced by the press 22 may then be physically transferred to a view box 24 for side-by-side comparison with the original image on the color monitor 11. Both the view box 24 and the color monitor 11 are preferably subjected to a standard lighting condition, as indicated by the broken line rectangle 26 in Figure 1. If the reproduction image does not accurately correspond to the original image, procedures as hereafter described may be used to make changes in the calibration data which are supplied to the computer 16 and to obtain a more accurate reproduction. once satisfactory reproductions are obtained and the reliability of the calibration data is established, reproductions of other original images may be generated.

The system is also arranged for storage of calibration data in a file along with an identification of the particular inks which are used, filed calibration data being thereafter retrievable for use by the computer 15 whenever the same inks are subsequently used. The filed calibration data may also be retrieved to provide initial calibration data when ink combinations are used which have not been used previously but which are similar to those identified in the file.

My aforesaid prior application, in Figure 2 thereof, shows how the monitor 11, computer 12, storage unit 13 and view box 24 may be physically arranged on a table and in a room or enclosure in a manner such as to provide the aforementioned standard lighting condition 26. Walls may preferably be used having a neutral color with flat, non- reflective characteristics and an indirect lighting arrangement is so used as to avoid reflections from the surfaces of a screen of the color monitor as well as from the view box 24 and a reproduction disposed therewithin. The lighting should be quite dim but sufficient to avoid excessive contrast between the screens being viewed and the reproduction and the background wall surfaces.

As is also disclosed in my aforesaid prior application, the view box 24 preferably has standard 5000K lamps which are variable in intensity using an intensity control knob and the computer 12 is connected to a keyboard and also to a status monitor. Computer 12 preferably includes disc drives for receiving transportable discs and it may preferably also include a hard disc drive. In addition, the computer 12 may be connected to a graphics tablet and associated puck, usable by an artist to develop a desired image on the screen of color monitor 11, using a standard type of "paint" software. A scanner may also be provided for scanning existing graphics and producing a corresponding image on the screen for editing by the artist as desired. A computer graphics arrangement of this type is very flexible and allows an artist to make changes easily and rapidly produce original images of a form desirable for reproduction.

1 In the practice of the invention, calibration programs may be loaded into the computer 12 for generation or editing of calibration data files which may be loaded and saved from and to a hard disc of computer 12 and/or transportable discs placed In disc drives thereof. During calibration, original test images may be developed on the monitor screen. for comparison with reproductions of such original test images which have been produced with the system of the invention and which are placed in the view box 24. The system of the invention facilitates an orderly calibration procedure such that errors are readily and quickly detected and corrected to directly converge to an accurate calibration for any particular combination of inks used in the reproduction process. Then the calibration data file may be stored and retrieved for use whenever that particular combination of inks is subsequently used.

Figure 2 is a flow diagram illustrating the digital processing operations performed by the computer 16 in the illustrated system 10. Initially, an original image file is loaded into the memory of the computer 16 from the storage unit 15 and is organized into sequential groups of pixel data, in an order corresponding to that required by the color processing system 20. Each group of pixel data contains sequential R, G and B groups of data which are 8 bit bytes in the system as described herein, each having a decimal value of from 0 to 255 and each defining the effective intensity of the corresponding component color of the pixel. It will be understood that the system is not limited to use of data in the form of 8 bit bytes and groups of data with a greater or lesser number of bits may be used.

Such numeric values of the pixel data bytes are referred to herein as the R, G and B values or i collectively as the RGB values. In the system as described herein and as is conventional, they have a complementary or inverse relationship to actual intensity, i.e., a zero value corresponds to a maximum actual intensity of the component color and a value of 255 corresponds to a minimum actual screen intensity of the component color. Maximum values for all three RGB values correspond to a black pixel and minimum values for all three RGB values correspond to a white pixel. If two of the RGB values are maximum values and the third is zero, the corresponding pixel is of one color at maximum saturation.

After loading of group of pixel data and testing for an end-of-file marker, gray factor and saturation factor determinations are made, for use in subsequent procedures to effect black printing, production and use of a color removal term, accurate tone reproduction, gray balance and color correction. It is noted that except as otherwise indicated, the procedures need not necessarily be performed in the order as indicated. For example, the black print procedure to determine the black output value may be performed after the procedures used to produce output color or chromatic values.

Gray & Saturation Factors GF & SF The first step in the illustrated procedures is the determination of gray and saturation factors. The gray factor corresponds to whichever of the red, green or blue component colors of a pixel has the lowest intensity, i.e., to whichever of the RGB values is the highest. It is used for control of black printing and also in determination of color removal and cleanness terms used in control of color printing.

1 t is The saturation factor corresponds to whichever of the red, green or blue component colors has the highest intensity, i.e. to whichever of the RGB values is the lowest. It is not used in control of black printing but is used in determination of cleanness and intensity terms used in control of color printing.

Black Printer In the operation as shown, the next step is to determine a black output value for printing of black ink. The black output value is a function of gray factor with 0 being the shadow (blackest black) and 255 being the highlight. A typical type of function is illustrated by line 28 in the graph of Figure 3. In this example, the black starts at a midtone value of the gray factor and steepens quickly toward the shadow. It is preferably developed using a look up table in the memory of the computer 15, in which the gray value is used to access a memory location which contains data defining the black output value.

Undercolor Removal/Gray Component Removal In control of color components which are developed as hereinafter described, a color removal value may also be developed as a function of gray factor. The color removal value CR corresponds to a ratio of undercolor removal (UCR) to gray component removal (GCR) and is developed as a function of the gray factor GF, typical relationships being shown in Figure 4. As shown, the value CR may be either a linear function of the gray factor GF as indicated by line 29, or a non-linear function as indicated by line 30, typically increasing from 0 at a midtone point to a value of on the order of 40% at the shadow end, as shown. As hereinafter described, the value CR is used to reduce each of the R, G and B values before tone reproduction values are produced therefrom.

Tone ReRroduction is Before discussing tone reproduction in detail, it is noted that in the illustrated system, output chromatic values are developed which are for control of cyan, magenta and yellow inks and which are designated as C, M and Y values or collectively as CRY values. In the illustrated system, each of such output chromatic values is derived, in part, from a tone value obtained from one of the RGB values and based upon an assumed approximate inverse relationship of the CMY ink colors to the RGB screen values and, in part, from a hue value developed separately from the RGB values, in a manner such as to obtain accurate tone reproduction, gray balance and color correction.

Before effecting tone correction, the R, G and B tone values are modified by adding the aforementioned color removal value CR, to obtain modified tone values R', G' and B' as indicated in Figure 2, the equations being as follows: R' = R + CR G' = G + CR B# = B + CR (CR is a function of the gray factor GF, typical relationships beihg shown in Fig. 4) Then further modified tone values Rl, G1 and Bl are determined as a next step in the system as illustrated. Tone reproduction is very important. No matter what is done to an image, if tone reproduction is not correct, the image will never look right. Tone reproduction is also dependent upon the characteristics of the device used for output of the image. The values R1, 1 i i 1 G1 and B1 are respectively functions of the R', G'-and B' values and may preferably be developed using a look-up table. Figure 5 is a graph in which lines 31, 32 and 33 illustrate typical relationships of the Rl, G1 and B1 values to R', G' and B' values.

With the aforementioned inverse relationship of the CHY colors to the RGB colors, the formulas now become: C = 255 - R1 M = 255 - G1 Y = 255 - B1 Gray Balance The second most important concept in color separation is the ability to reproduce gray through the tone scale from highlight to shadow. For each set of colors used to reproduce with inks, dyes or other media, there is a set of values for each that will produce gray at each point on the tone scale. If each output color (e.g. cyan, magenta or yellow) were pure, then equal values would make gray. This is not usually the case and the colors will be contaminated to some extent.

Generally, it is advantageous to determine the values of magenta and yellow as a function of cyan because cyan is usually the most contaminated and most always will be of the highest value. In the illustrated system, the formula for the cyan value remains the same, but new G2 and B2 values are developed, as functions of the G1 and B1 tone values for subtraction from 255 to obtain values for control of the magenta and yellow values.

In the graph of figure 6, lines 34 and 35 indicate that for control of magenta and yellow, the ratios of the G2 and B2 values to the corresponding G1 and Bl values are typically less than unity. Line 36 1 represents a unity ratio between ordinate and abscissa and corresponds to the unchanged Rl tone value used for control of cyan. It should be understood that although the straight lines indicate that the functions are linear, they may be non-linear and may be generated through lookup tables or the equivalent. The formulas now become:

C = 255 - Rl M = 255 - G2 Y = 255 - B2 Color Correction In color separation, color always needs to be independent of gray and vice-versa. In order for this to hold true, any term used to correct color by definition must drop out in pure gray. In accordance with a very important feature of the invention, color is corrected based upon a hue value for each pixel, representing only a color angle and being independent of gray. In particular, a hue value H is developed through formulas which are based upon an assumption that a red vector is at a 0 angle, a green vector is at an angle of 2r/3 radians (120 degrees) and a blue vector is at an angle of 4r/3 radians (240 degrees).

With the R value as a reference, an equation for Hue is as follows: H = 90 - [180(rad/w)] if G is less than B: H = H + 180 where rad is in radians: rad = Arctan (f/square root of 3) and where f = [(2R - G) - B]/(G - B) When terms are combined and the Arctan is expressed in degrees, the formula is as set forth in Figure 2.

z 1 To define a more complete color correction equation, additional terms are used. The gray factor GF and the SF factor have previously been defined as respectively the highest value of RGB and the lowest value of RGB. Another term is called Grayness and represents a percentage of grayness of a given set of RGBs:

Gr = (255-GF)/255 - SF) Cleanness is the inverse of Grayness:

Cl = 1 - Gr The last term is Intensity which represents the percentage depth of color:

I = 1 - (SF/255) Color separation equations can now be written as:

cyan = 255-Rl(l + some function of I, Cl, H) Magneta = 255-G2(1 + some function of I, Cl, H) Yellow = 255-B2(1 + some function of I, Cl, H) The term (some function of I, Cl, H) must drop out at perfect gray (R=G=B). At perfect gray GF = SF, therefore Grayness = 1 and Cleanness = 0. The term therefore drops out. The required functions of H are typically non-linear and are established through look-up tables or the equivalent. Functions for I and Cl may similarly be established although linear functions may be used in some cases depending upon the accuracy which is required or desired. The function for H is either positive or negative, depending upon the need to add or subtract color.

i is The general equations are as follows:

c = 255 - R1[1 + ICCItHc] M = 255 - G2[1 + Imci4%] Y = 255 - B2[1 + IyavHY1 where Ic, IN and Iy are functions of Ir CLc, CIq and CI are functions of Cl and Her IIN and Hy are functions of H.

It has been found that in actual tests with certain available inks, satisfactory results can be obtained when each of the Ic, IN and Iy terms is made equal to I and when each of the C1t, CIL, and CL, terms is replaced by CL? so that the following simplified equations may be used successfully in appropriate circumstances:

C = 255 - R1[1 + Ici 2 He] M = 255 - G2[1 + IC12H,'] Y = 255 - B2[1 + IC12H Y] In applying either the general or the simplified equations, the Her HV and HY look-up tables are developed in an iterative fashion using calibration procedures of the type disclosed in detail in my aforesaid prior application. Test targets are output for color proofs which are evaluated against the computer display image, the lock-up values being changed until satisfactory results are obtained.

With the system of the invention, the calibration procedures are greatly simplified and can be much more quickly accomplished than would otherwise be the case. This results from the fact that each operational step and the look-up or other control values associated therewith have an effect on the proof which can be determined visually and which, so far as possible, is substantially independent of the effect of control values for other operational steps. In other words, there is no overlap of corrections, and calibration procedures converge to rapidly obtain results which are as visually accurate as is desired.

The generation and use of the Hue value is particularly important because of the importance of correct hue in obtaining a reproduction which is visually accurate. With the system of the invention, differences in hue between an original and a proof can be readily detected visually and appropriate changes in control values can be determined and made.

The H look-up tables are preferably generated in 1 degree increments of Hue, thus creating 360 discreet values. A sample diagram of the correction coefficient versus Hue is shown in Figure 7. Straight lines are shown, but it will be understood that in practice, smooth curves are generally used. The diagram of Figure 7 is for the Cyan printer. The same diagram with different curves is also applied for the Magenta and Yellow printers.

It should be understood that control values may be used instead of or in addition to those included in look-up tables for defining the various functional relationships as disclosed herein, and it will be understood that other modifications and variations may be effected without departing from the spirit and scope of the novel concepts of the invention - 18

Claims (21)

CLAIMS:

1. A method for reproduction of images. comprising the steps of supplying original image digital data in the form of groups of input data each of which defines R, G and B values variable within a certain range and respectively corresponding to intensity values of red, green and blue component colors of an individual pixel of an original image, and digitally processing said groups of input data to produce from each group of input data a group of output digital data providing output values including at least three output chromatic values for control of reproduction of an individual pixel of said original image in at least three output colors, said processing of said groups of input data including the steps of producing from said R, G and B values a hue value corresponding to the-hue of said individual pixel of an original image, and using said hue value in producing said output chromatic values.

2. A method as defined in claim 1, wherein said hue value is so produced as to lie in a range of from 0 to 360 degrees and corresponds to an angular rotation from a certain reference hue.

1 - a

3. A method as defined in claim 1, wherein said producing of said hue value includes the steps of producing a value equal to the arctangent of the ratio of the sum of twice the value of one of said R, G and B values less the other two of said R, G and B values to the product of the square root of 3 and the difference between said other two of said R, G and B values, and subtracting said arctangent value from a constant value and adding either 0 degrees or 180 degrees depending upon whichever of said other two of said R, G and B values has the higher value.

4. A method as defined in claim 1, wherein said processing of said groups of input data includes the step of producing from said R, G and B values an intensity term related to the intensity of any net color component of said individual pixel of an original image, and using said intensity term together with said hue value in producing said output chromatic values.

5. A method as defined in claim 4, wherein said intensity term corresponds to the highest of the intensity levels represented by said R, G and B values.

1 - 20

6. A method as defined in claim 1, wherein producing of said output data includes the steps of producing a cleanness term related to the ratio of the highest of the intensity levels represented by said R, G and B values to the lowest of the intensity levels represented by said R, G and B values, and using said cleanness term together with said hue value in producing said output chromatic values.

7. A method as defined in claim 1, wherein producing of said output data includes the steps of producing first, second and third tone values as predetermined functions of said R, G and B values, and using said tone values together with said hue value in producing said output values.

8. A method as defined in claim 7, wherein producing of said output data further includes the step of producing from said R, G and B values an intensity term related to the intensity of any net color component of said individual pixel of an original image, and using said intensity term together with said tone values and said hue value in producing said output chromatic values.

1 1

9. A method as defined in claim 7, wherein producing of said output data further includes the steps of producing a cleanness term related to the ratio of the highest of the intensity levels represented by said R, G and B values to the.lowest of the intensity levels represented by said R, G and B values, and using said cleanness term together with said tone values and said hue value in producing said output chromatic values.

10. A method as defined in claim 1, wherein said output values include a black output value in addition to said at least three output chromatic values.

11. A method as defined in claim 10, wherein said processing of said groups of input data include the step of producing a gray factor corresponding to the lowest of the intensity levels represented by said R, G and B values, and using said gray factor in producing said black output value.

12. A method as defined in claim 11, wherein producing of said output data includes the steps of producing first, second and third tone values as predetermined functions of said R, G and B values, producing first, second and third modified tone values as functions of said gray factor, and using said modified tone values together with said hue value in producing said output values.

13. A method as defined in claim 11, including the step of producing a color removal term as a predetermined function of said gray factor, and adding said color removal term to each of said first second and third tone values in producing said first, second and third modified tone values.

14. A method as defined in claim 11, wherein said using of said modified tone values includes the steps of producing first, second and third further modified tone values as predetermined functions of said first, second and third modified tone values, respectively.

15. A method as defined in claim 14, wherein said using of said modified tone values further includes effecting a gray balance correction by producing at least two additionally modified tone values as predetermined functions of at least two of said first, second and third modified values.

1

16. A method as defined in claim 1, wherein producing of said output data includes the steps of producing first, second and third tone values as predetermined generally inverse effective functions of said R, G and B values, and using said first tone value together with said hue value in producing a cyan output chromatic value, using said second tone value together with said hue value in producing a magneta output chromatic value and using said third tone value together with said hue value in producing a yellow output chromatic value.

17. A method as defined in claim 16, including the steps of producing first, second and third hue terms as predetermined functions of said hue term for correcting said tone values in producing said cyan, magenta and yellow output chromatic values.

18. A method as defined in claim 17, including the step of producing from said R, G and B values an intensity term related to the intensity of any net color component of said individual pixel of an original image, and using said intensity term together with said first, second and third hue terms in producing said cyan, magneta and yellow output chromatic values.

- 24

19. A method as defined in claim 18, wherein said processing of said groups of input data includes the step of producing a gray factor corresponding to the lowest of the intensity levels represented by said R, G and B values, using said gray factor in producing a black output value, and producing first, second and third modified tone values as predetermined functions of said gray factor for use with said hue and intensity terms in producing said cyan, magenta and yellow output chromatic values.

20. A method as defined in claim 19, wherein each of said predetermined functions is modifiable independently of others of said predetermined functions for calibration purposes.

21. A method for reproduction of images substantially as described herein with reference to the accompanying drawings.

1 Published 1991 at Ile Patent Office, State House. 66171 High Holborn. London WC1R47P. Further copies may be obtained from Sales Branch. Unit 6, Nine Mile Point. Cwmfelinfach. Cross Keys. Newport. NPI 7HZ. Printed by Multiplex techniques lid. St Mary Cray. Kent.