US2176518A - Photomechanical color process - Google Patents

Description

Oct. 17, 1939. c u 2,176,518

I PHO'IOMECHANICAL COLOR PROCESS Filed Dec. 30, 1937 Fig.1

Fig.2.

COLOR BLUE GREEN ORIGINAL ig. ,,4, P/B/O/B 4/27"- PARTLY CORRECTED BLUE MADE NEGAT/ VE PR/N TING EXPO5URE I! PARTLY CORRECTED POSITIVE FOR YELLOW PRINTER SEPARATION NEG'A Tl VE ORIGINAL PLUS 0ME GREEN BLUL-TI' GRIT/w SEPARATION NEGATIVE CORRECTED BLUE MADE GREEN NEGATIVE SEPARATION NEGATIVE PRINTING EXPOWJFE CZFRECTE POST/Vt OIN! ACYUEL? FOE YfLLOF V PRINTER INVENTOR W IYWM/ BY f.

ATTORNEYS Patented Oct. 17, 1939 UNITED STATES PATENT OFFICE' 2,176,518 PHOTOMECHANICAL COLOR P ROCESS ApplicationDecember 30, 1937, Serial No. 182,503

6 Claims.

This is a continuation in part of my patent application Serial No. 138,350 which relates to photo-mechanical process work and more particularly to the color filters employed in the making of color separation images of a multi-colored original for three or four-color processes.

In order to overcome the deficiencies in even .the best contemporarily available colored inks employed in photo-mechanical process work, it is customary to introduce color correction by means of hand retouching or so-called masking method of color correction.

It is an object of the invention to provide color separation negatives and a method of making them so that they will be particularly suitable for use in a process including masking. According to my invention, therefore, the abilities of the masking method will be more fully realized than heretofore.

.It is thus the object of this invention to provide a method of making a color separation negative which, in the sense that it gives better color correction in the final print, is more highly corrected than hitherto.

It is a further object of the invention'to provide a more highly corrected color printer than hitherto.

The invention consists essentially in an improvement of, or more exactly a correction of, "the masking method of color correction. By way of example of the invention, a blue filter negative which is intended for use in a masking process wherein it is to be masked by a thin positive-from the "green filter negative (or in some cases two positives corresponding to both green and red) is not made according to the prior art by exposing only to blue light from the original,

' .but is exposed to blue plus some green (or blue apparent from the following description when read in connection with the accompanying drawing which shows:

Fig. l. A color chart representing three greatly magnified halftone dots made with three typical colored printing inks.

Fig, 2. A filter made according to one form of my invention. 1

Fig. 3 is a flow chart indicating the making of a yellow printer by the masking method according to the prior art.

Fig. 4 is a flow chart of the invention as applied to the example shown in Fig. 3.

In three-color photography and photo-mechanical process work for reproducing a multicolored original, it is customary to make three color separation images by exposing three photographic layers to the primary colors reflected or transmitted by the original. The three primary. colors, red, green, and blue, cannot and need not be exactly defined but are well known to those skilled in this art. The three corresponding subtractive or complementary colors, blue-green, magenta, and yellow are sometimes called blue,

red, and yellow but this latter nomenclature and its accompanying confusion will be avoided in this specification. Some color processes employ three superimposed photographic layers; others use three separate films or plates. The various techniques are Well known and need not be listed here. In so-called additive processes, color separation positives of the three primary colors are placed substantially in register by optical projection or by juxtaposition of elemental areas thereof such as in a color screen, process. The present invention relates generally to the socalled subtractive processes which employ color separation positives in the three subtractive colors, yellow, magenta, and blue-green, and more particularly to the processes which employ three inks or similar coloring materials. The invention is particularly useful when a reproduction is made to be viewed by reflected light such as a print on paper.

For clarity, the following description will be confined to so-called halftone processes wherein the printing density at any point is determined by the effective area of some elemental portion such as a halftone dot. The term printing density includes both continuous tone density and halftone density, the latter being related to the effective, printing area of individual halftone dots. The term ink refers to any coloring material of the types commonly used in the reproduction of a multi-colored scene, painting or color photograph.

The accompanying color chart Fig. 1 represents three greatly enlarged halftone dots'which overlap somewhat. A halftone image made up of a large number of such combinations can be considered as comprising a certain relative distribution of eight colors, or color patches as shown, namely, the inks yellow, magenta, and blue-green," the areas where two dots overlap,

red, green, and blue, the areas where three dots overlap, black, and the white of the support on which the picture is printed in reproduction. As is well known, due to deficiencies in the inks this blue appears purple and the black appears brown. The refiectivities (the ratio of the reflected light to incident light of any given color) corresponding to the eight colors will be referred to in this specification and the accompanying claims as the reflectivities of the sep-.

arate coloring materials and the reflectivity of the combined coloring materials and the reflectivity of the white support. Reflectivity depends on the color (i. e. the spectral distribution of energy), the incident light and the spectral sensitivity of the eye or photographic plate measuring the light intensities as well as on the properties of the coloring material being tested. Since the invention concerns only photographic rather than visual refiectivities, the phrase reflectivity to primary red means reflectivity, measured with a suitable photographic emulsion, for red light defined by the transmission of a so-called red filter illuminated with white light.

Since the colors of even the best contemporarily available yellow, magenta, and bluegreen inks are not exactly minus-blue, minusreen, and minus-red, a certain amount of color correction by hand retouching or by masking is necessary if good color rendition is to be obtained.

As is well known, one common form of the masking method of color correction comprises making a color separation negative through one of the primary filters and superimposing in register on this negative, a mask consisting of a low contrast low density positive made from a negative taken through another filter which transmits one or both of the other primary colors. For example, the negative made through the primary red filter needs little if any masking; the negative made through the green filter may be masked with -a positive corresponding to the negative made through the red filter; and the negative made through the blue filter may be masked with a positive whose negative was made through the green or a yellow filter. Although the color rendition when masking is employed, is much better than that in which no color correction is applied, it is still not perfect.

As an example of this form of the masking method, Fig. 3 in the form of a flow chart, shows a negative made through a blue filter and masked with a thin positive of a negative made through a green filter. The chart is more or less self explanatory. The letters R, Y, G, BG, B and M near the original refer of course to the colors red, yellow, green, bluegreen, blue and magenta. The arrows indicate exposures which may be by projection or contact. The heights of the shaded areas in each box are very roughly indicative of the corresponding image density, when the original colors are of equal density which in practice rarely occurs.

According to my invention better correction and more improved color rendition is provided by a method which employs a special set of filters instead of the usual three primary color filters in exposing the negatives. These special filters are constructed of such materials and/or employed in such a manner that they give color separation images particularly adapted to be masked with images made through other filters and to give color printers particularly suitable for a specified' set of inks. The above mentioned patent application of which this is a division, relates to these special filters and the method of manufacturing them.

The present application relates specifically to the novel form of the masking method which constitutes the invention. According to the invention when making a color separation negative to be masked by a positive corresponding to a different color from the original, the negative is exposed not only to its own color from the original but also to a portion of this different color which corresponds to the mask.

In the example shown in Fig. 4, the bluefilter negative is exposed not only to blue light from the original, but also to a portion of some other color which, since the blue made negative is to be masked by a "green-filter positive. happens to be green. The exposure of this bluefilter negative may be made through special filters as discussed in the above mentioned copending application or successively exposing to blue and green in predetermined proportion.

According to a simple theory, any two coloring materials, such as two inks should, in combination, have a reflectivity which is the product of the separate refiectivities with respect to any given color. Practically, however, this is not the case and because of this, the ordinary masking method of color correction is not able to utilize the full abilities of available inks. This discrepancy between the reflectivity of the inks in combination to any given color and the product of the refiectivities of the separate inks to this color, will be discussed in greater detail below.

My invention will be more clearly understood from the following outline of one method of computing, from the primary color reflectivities of the particular inks which are to be used, the desirable spectral transmission characteristics of the color filters through which the color separation negatives are to be made and the amount and kind of masking to be employed. For purposes of the computation, the eight colors corresponding to the inks to be used in the reproduction of a multi-colored original are mounted beside the original, using the actual inks themselves and their white support. It is a natural and approximately true assumption that the best possible reproduction of the original will be obtained when those eight colors are exactly reproduced. Methods of making color printers according to my invention will be included in this sample computation.

colors respectively. The corresponding reflectivities are given in the columns Rb, Rg, and Rt- Perfeet R R print x y ing Ink pamhcs Db R Db .6D .85Rb |85b+ 15R D; D; .6D; 8Rb+2Rz D,- Dy .68D densities 1.50 Yellow" 11.0 .96 (1.50) 9.4 24.4 .61 .61 (1.50) 28.8 .54 .54 (1. 50) 1.50.". Y+M (rt 5.4 .58 (.90) 4.6 5.7 1.24 .55 (1.35) 5.7 1.45 .46 (1.28) Ma enta 22.9 .04 (.06) 19.5 21.1 .68 .08 (.20) 20.3 .99 .01 (.03) 0 M and 11.2 .07 (.11) 9.5 10.1 1.00 .12 (.30) 9.7 1.01 .02 (.06) 0 Blue green .40 30.8 .03 (.05) 33.8 34.3 .46 .03 (.07) 35.7 .45 .03 (.08) 1.50. BG+Y (creel-1).." 1.20 6.3 66 (1.03) 5.4 7.3 1.14 .60 (1.48) -7.6 1.12 .51 (1.41) 1.50.... BG+Y+M (black). 1.44 3.6 .54 (.85) 3.1 "3.6 1.44 54 (1.33) 3.6 1.44 41 (1.22) 0 White 100 (0) 85 100 0, 100 0 (0.

r R! D R; .6D D,.62D. R: .BR, 8R+ 2Rr DI D,.66Dr

. 0 0 100 0 0 (0) 15.C 80 100 0 0 (0) 1. 1.14 7.2 .69 1.08 (1.50) 1.08 5.8 22.1 .66 .60 (1.50) l.- 1.00 10.0 .60 .97 (1.35) 1. 50 8.0 25. .60 56 (1.40) l. 1.46 3. 5 .88 .60 (.83) .52 2.8 3.6 1. 44 .53 (1.33) 0 Blue green... .71 19.5 .45 .02 (-.03) 2.92 15.6 17.0 .77 .00 (9) 0 BG-l-Y (green) .89 12.9 .53 .03 04) 1. 94 10.3 11.1 .90 .01 (--.02} 1.50 BG+Y+M (Perfect) Ink patches Dr R. 1.071)r .62Dr .2R. .66Dr

Yellow 0 100 0 0 0 Y-l-M (red) .09 81.5 .1 .06 16.3 .06 Magenta .06 87.0 .06 .04 17.4 .04 M+BG(bluc) 1.38 4.2 1.48 .86 .84 .91 Blue urcen 1.17 6. 8 1.25 73 1. 36 77 BG+Y(ETO0I1). 1.38 4.2 1.48 .86 .84 .91 BG-l-Y-l-M (bla 1.40 4.0 1. 50 .87 .80 .92

The above table shows a computation made from the actual densities of the eight colors corresponding to three typical inks and with respect to three primary colors as defined by three primary color filters which I chose for the purpose. The red filter is that known as Wratten Light Filter No. 25. The green filter is that known as Wratten Light Filter No. 61 and the blue filter is that known as Wratten Light Filter No. 47. These particular filters were chosen to meet the general requirements that a primary color filter shoul'd sharply define a certain known portion of the spectrum. Similar computations can be made for any similar inks and set of filters. The letter D as it appears at the head of certain columns refers to density. The letter R similarly refers to reflectivity, and the subscripts refer to the filter through which the density or reflectivity was measured. For instance Db, D; and Dr are the densities measured through the #47, #61 and filters respectively.

Since the invention is concerned only with photo mechanical processes, these densities were measured photographically using film or plates having the spectral sensitivity of the material with which the separation negatives are to be made.

Using the white support upon which the multicolored original is to be reproduced, I printed a group of overlapping color spots and measured the refiectivities of the separate coloring materials and the coloring material in combination, with respect to the three primary colors as defined by the above listed filters. The actual density readings obtained are indicated in the columns Db, D and Dr for the three primary Following the usual conventions, I use the terms absorption to mean 1-R (reflectivity expressed as a fraction) and opacity to mean l/R. Density is the logarithm of opacity. For the purposes of this typical computation, I am considering only the straight-line portions of the characteristic curves of photographic emulsions. In the accompanying claims all of the well known effects of photosensitivity characteristic curves are fully included by the term photosensitivity function which covers the efiect of the toe and shoulder of the curve as well as the purely logarithmic factor introduced by the straight line position. When a photographic emulsion is developed to a low contrast the relative densities of the image are reduced proportionally; such effects are indicated in the above table by multiplying the densities by a factor equal to the photographic contrast. In one form of the masking method of color corrections, as discussed above, a positive is made from the negative which is taken through the primary green filter and this mask is processed to be of low contrast. say \/=.6. For purposes of computation, the effect of masking a negative taken through a blue filter with a positive corresponding to the green filter. when employed with those particular inks can be represented by subtracting corresponding 7 densities as shown in the column Db.6Dg. By increasing the development when making a photographic image, the relative densities of the various portions are increased proportionally. ,For purposes of comparison, the computationsmade in the above table are corrected to a maximum density of 1.50 by multiplying by a suitable factor (corresponding to developing to a suitable contrast), as indicated by bracketed numbers accompanying the various density columns.

The term photosensitivity function defined above, covers all such variations in contrast and other effects on relative tones introduced by one or more photographic steps. It is thus a particularly convenient term since it describes in two words the mathematical factors introduced by the steps of this general type of process which are those common both to the invention and the prior art.

The eight colors on the chart mounted adjacent to the original will reproduce perfectly if the amount of each ink in the reproduction is equal to the amount of the same ink in the corresponding point on the chart. This condition is obtained with a perfect printer having densities to the three primary colors as indicated in the column at the extreme left of the above table.

It is obvious that even when employing the masking method, this theoretically perfect ratio of the effective densities is not approximated very closely, as shown by comparison of the Db.6Dg column with the perfect printing densities.

I have found that the correction obtained by the masking method would be approximately perfect if the refiectivities of the coloring materials in combination equaled the product of the refiectivities of the corresponding coloring materials separately. For convenience in the accompanying claims, I refer to this as the product law" of the eight refiectivities. It will be seen from the above table that this is not so, namely Rb for the red patch is not equal to Rb for the yellow patch timesRb for the magenta patch (i. e., Db yellow plus Db magenta is not equal to Db red).

According to one form of my invention as illustrated in general by Fig. 2, I provide a bluish filter which transmits primary blue and primary green in the ratio of 85 to 15. In the above table I refer to this filter as X. As shown in the table, the reflectivities of the eight colors to the color of the X filter equals .85Rb+.15Rg. The corresponding densities are given in the column D};- It will be noted that the discrepancies between the refiectivities of the coloring materials in combination and the product of the reflectivities of the separate coloring materials is appreciably reduced.

The units expressing the amounts of the primary colors as defined by the filters chosen, are such that equal quantities haye equal effect on the panchromatic emulsion used. It will be obvious that such units take care of the filter factors of the emulsion.

If a color separation negative were made through an X filter, masked with a .6Dg positive as before and employed with these particular inks, the degree of correction is indicated in the column Dx.6Dg. It will be noted that the figures in this column are approximately proportional to the perfect printer densities and also that my invention has thus provided a method of color reproduction which gives a much better rendition of the eight colors mounted by the original than that obtained by the masking method when the original color separation negative is made through a primary color filter. This is due to the fact that the effective densities of the colors requiring a solid tone of yellow ink in their reproduction are high and equal.

Because of numerous variations which appear in the practice of any photo-mechanical process,

mary colors.

arrears it is sometimes not worth while to reduce the deviations from perfect much beyond that shown in this Dx-.6Dg column. However, the effect of a filter Y which transmits primary blue light and primary green light in the ratio of 4 to 1 is given for purposes of comparison. In practicing my invention I prefer to repeat my calculations using two or three ratios of the two pri- It is not necessary that both of these colors be primary colors, but I have found that the best results are obtained when the color which is added to the primary corresponding to the printer being made is the same color as that of the filter corresponding to the mask used in making this printer. For example, if the blue I filter negative were to be masked with a yellow filter positive instead of a green filter positive as described above, the bluish filter used in making this negative should, according to the preferred embodiment of my invention transmit blue and yellow in the proportion of say, 8 to 1.

According to a different embodiment of my invention, each color separation negative is made by successively exposing a photographic film or plate to a primary color reflected from the original and then to the color reflected from the original corresponding to the mask which is to be used with that negative. For example, in the first case shown in the above table, the blue" filter negative is exposed through a blue filter for a certain time and then through a green filter for a relatively short time; the times being in the ratio 85:15, if the filter factors of the emulsion for the blue and green filters are equal. The ratio is modified by the ratio of these filter factors when they are unequal. By this method, a color separation negative is obtained capable of utilizing the full benefits ofthe masking methd of color correction.

The above table includes similar computations for the green and red filter negatives. In the ease of the green filter negative a Z filter transmitting green and red in the ratii. of 4:1 is shown to be appropriate for these particular inks and to give a negative suitable to be masked by red filter positive of .66 contrast. In the case of the red filter negative, the figures in the 1.07 Dr column indicate thatthe eight colors for these inks have an approximately correct effective density to red light and hence masking is not necessary.

It will be noted by those skilled in this art that the above tabular method lends itself particularly well to the making of these computations, especially since certain approximations are allowable and in some instances cannot be avoided. Once a set of color filters has been computed as described above, no further computations are graphic emulsion by which their refiectivities were measured. Furthermore many printing inks commercially available at present differ only to a small extent and hence the same set of filters is useful with various commercial sets of inks.

It is customary in the making of gelatin color filters, to determine the desired spectral transmission of the filter and then to choose (from experience and by tests dictated by this experience) a suitable coloring material which will impart to the gelatin this particular spectral transmission.

Having thus described methods of making color separation negatives particularly suitable to be masked and a set of filters made in accordance with my invention from the refiectivities of the inks to be employed, I wish to point out that my needed when using this set of inks and the photo- 'method of making a color printer which comprises exposing a sensitive photographic layer to one of the primary colors from the original and to a different color from the original, developing the layer, preparing a positive image corresponding to a color from the original effectively the same as said different color, making a negative comprising said developed layer masked by said positive and printing a positive printer therefrom.

2. In a photomechanical process employing three subtractive-color coloring materials for the reproduction of a multicolored original, the method of making a negative which comprises measuring the primary color refiectivities of the eight colors corresponding to the support upon which the reproduction'is to be printed, the coloring materials and the combinations thereof, computing therefrom a color distribution comprising one of the primary colors and a relatively small amount of another color such that the readapted to be masked by a suitable positive corresponding to a color from the original which comprises measuring the primary color densities of the eight colors corresponding to the support upon which the reproduction is to be printed, the coloring materials and the combinations thereof, computing therefrom a color distribution comprising one of the primary colors and a different color approximately the same as that corresponding to the masking positive such that with respect to this color distribution, the effective densities, when masked by said positive, of those of the eight colors requiring for their reproduction the coloring material complementary to said primary color are high am approximately equal and the similar densities of the remainder of the eight colors are approximately equal to zero, exposing a sensitive photographic layer to light from the original having this color distribution and developing the layer.

4. In a photomechanical process employing yellow magenta and blue-green pigments for the reproduction 01'. a multicolored original having actually or hypothetically associated therewith eight color patches including the three pigments individually and in combination and the support upon which the reproduction is to be printed, the

' method of making a yellow printer which comprises making a color separation negative of the original for a color having two components C and D, where C is primary blue and D is a correcting color, making a second color separation negative of' the original for a color substantially consisting of D only, making a thin positive mask from this second negative, masking the first negative therewith in register and printing a corrected positive from the combination of the nega-,

tive and mask, the ratio of the amount of C to the amount of D in the color having two components being such that in a similarly corrected positive of the eight color patches the apparent densities of those including the support alone which do not contain the yellow pigment are low and approximately equal and the apparent densities of the other patches are high and approximately equal.

5. In a photomechanical process employing three subtractive color coloring materials for the reproduction of, a multicolored original having actually or hypothetically associated therewith eight color patches including the coloring materials individually and in combination and the support upon which the reproduction is to be printed, the method of making a printer for one of the coloring materials which comprises making a color separation negative of the original for a color having two components A and B where A is the primary color substantially complementary to said one of the coloring materials and B is a correcting color, making a second color separation negative for a color substantially consisting of B only, making a thin positive mask of predetermined contrast from this second negative, masking the first negative therewith in register and printing a corrected positive from the combination of the first negative and the mask, said predetermined contrast and the ratio of the amount of A to the amount of B in the color having these two components being such that in a similarly corrected positive of the eight color patches the apparent densities of. the patches not containing said one of the coloring materials are low and approximately equal and the apparent densities of the four patches containing said coloring material are high and approximately equal.

6. In a photomechanical process for reproduc ing a multicolored original, the method of mak- I said correcting color and photographing the color separation negative masked by said positive.

JOHN A. C. YULE.

CERTIFICATE OF CORRECTION. Patent No. 2,176, 518. October 17 1959.

. JOHN A. C. YULE.

It is hereby certified that error appears in the printed specification of the above numbered patent requiring correction as follows Page 5 line 6, in the heading to the seventh column of the t able, for .8 5 read .85R and that the said Letters Patent should be readwith this correction therein that the same may conform to the record of the case in the Patent Office.

Sigaed and sealed this 2nd day of January, A. D. l9hO.

Henry Van Arsdale Acting Commissioner of Patents.

Images