US2244992A - Color responsive correction method for polychromatic reproduction - Google Patents

Description

June 10, 1941.

s. 5. YGUERRERO COLOR RESPONSIVE CORRECTION METHOD FOR POLYCHROMATIC REPROISUCTION Filed Nov. 8, 1938 2 Sheets-Sheet 1 INVEN TOR Av I A r w d x 4 J 2 l nmvarz:

6'. 6141mm Gus/mam 6 6 3 2 a IA Ey- MAslrs June 10, 1941. GT GUERREfiQ 2,244,992

COLOR RESPQNSIVE CORRECTJiON METHOD FOR POLYCHROMATIC REPRODUCTION Filed Nov. 8, 193a j 2 Sheets-Sheet 2 [915?56750 15141 AVA/CED IGJITIVES I IN V ENTOR.

3. 5'. GA/Pc/A 6049mm,

Patented June 10, 1941 COLOR RESPONSIVE CORRECTICN METHOD I FOR POLYCHROMATIC REPRODUCTION Salvador Garcia Guerrero, Mexico City, Mexico Application November 8, 1938, Serial No. 239,547

1 Claim.

My invention relates to natural color reproduction by photographic and photo-mechanical,

means and is particularly directed to overcoming the inherent deficiencies in color characteristics of the media customarily employed. As willg'lbe color compensation.

complished.

sive to each color fault.

stimulation is added.

blue, C, which reach the eye.

filters.

blue taken of blue light.

tives are made, thus providing records which indicate the relative degree of absorption necessary I to produce the original scene gradation in each color.

, To control the red stimulation, the positive of the red record may now be printed. or otherwise and red- The C or yellow pigment most 5 seen, this is done in an objective rather than colored in such way that red light is absorbed in subjective manner, not liable to thejudgment of the proper degree and distribution; However, the operative, by photographic color responsive this red absorbent or minus-red color must control only red light, not having any efi'ect on The invention eliminates the tedious hand reblue or green light, for complete control or blue touching and staining methods which are commust be responsive to the blue-light-responsive monly used. Such methods of correction depend positive; and absorption of green be controlledon the color judgment of the workman, and the solely by the green made positive. Therefore, result is-merelyhis guess-work. For this slow, by the same reasoning, the green light is conexpensive and inexact procedure, substitute 1 trolled solely by a minus green" density in its an exact, scientifically controlled and corelated positive record; and blue light is modulated enprocess which permits rapid and accurate .work tirely by minus blue pigment in its record. to be done and a greater volume of work ac- Now when the effect of these color controllers are synthesized, by superposition or other suc- The whole reproduction sequence is controlled cessive action on white incident light, each conby exact measurement, the faults inherent in the trols its own color sensation, not infringingon inks are automatically recorded and proper corany other, and the relative distribution of the rection is calculated from measurements responthree primaries is reproduced as it was in the original scene. I For a definite understanding of the theory 2 The colors minus red, "minus green," and underlying my process a brief explanation of the "minus blue are called the subtractive primaries subtractive process of color reproduction is helpor printing primaries and. might properly be ful. It has been learned by experiment that termed blue-green, blue-red," human vision depends on the stimulation of three green, or also cyan, magenta and yellow. Howresponse mechanisms of the nervous system, one ever. e customary e ms are blue. red/7 being stimulated by reception at the eye-of blueyellow, and they are conventionally referred violet light, another by green light, and the third to as A, B and C, which characters also by red-orange light. The color sensation is dem y refer to the takingfilters- For convenience termined by the relative stimulation or these the terms A, B, and crwillbeusedtodesignate the three responses by the light being observed. For separations; A being taken through a red filter example, it all areequally stimulated, the imand printed in blue ink; B being taken'through pression is white; if red is lacking the impression a green filter and printed in "red ink, and C due to equal stimulation by the blue and green being ken through a e filter d p ed in is blue-green, which becomes more white as red yellow ink.

40 The faults in practica'lsj-application oi the above Then to control the color sensation, it is necesh ry in h fact that h P n u d in sary and m t t trol the relative intenprinting absorb the colors which they should not sity of each of these three visual, or additive, absorb and that the taking filters transmit colors primaries, commonly called red, A, green, B, and which they S o not transmit, thus controlling not only the color each is designed to modulate The scene is therefore analyzed to determine but also affecting the intensity o the other p the relative intensity of these primaries bysucmaries- That the A Pigment is e gned to cessively photographing through like colored absorb and Control y ed light, but i also a This results in separation negatives; the sorbs blue and green, which it should reflect, or opacity of that exposed to red light being an in- 5 transmit, perfectly. The B pigment should abverse record of the intensity and distribution of sorb only green l ht, but it also has considerable red light; the green.taken of green light, the density to b enearly perfect, having very little absorption of For convenience, positive prints of these nega- 5 red or green light. 5

My process provides integrated control and v measurement of these faults and provides'for their accurate compensation.

Further objects,- advantages and features of invention may be apparent in the following de-.

scription and accompanying drawings, wherein time required for development of the mask at standardized conditions.

on each side of the field l2, so that they will be reproduced on the edges of the negatives and may be readily trimmed from the final correctedpositives or printing plates.

1 Separation negatives l3, of equal contrast, at about gamma equal to one, are made in the usual way, thru A, B, and C filters; and a grey printer negative is exposed thru yellow filters of different absorption bands, or thru infrad-red filters, or by successive balanced interposition of the tricolor filters, or from a combination diapositive from the A, B and C separation negatives.

The gradation of the respective step strips and their various reproductions are graphically shown in Figure 3 by the use of the convention that the solid density, of the darkened wedge, lying in a step is evenly distributed over the area of that Figure 6 shows a measuring instrument applied to a separation negative.

In my color reproduction process the filters used should approximate as nearly as possible the theoretical standard and the pigments are chosen as nearly as possible to beeach the complementary of its taking filter as this permits brighter tints to be attained in correct hues. However, it is necessary to adjust and balance the inks so that they have the proper relative refiective characteristics, and so that these are properly tied into the entire sequence of reproduction.

This is accomplished by the use of a color disc [0, which comprises three equal, 120, sectors as shown in Figure 1a. Using printing plates of the same qualities and which are made under reproducible conditions, one-sector is printed in. A ink, blue green; one in B ink, magenta; and

step. A further convention followed is that faults are represented. by central white wedges, while purposeful densities are represented by central black wedges.

It may be seen however, that the negative I3A exposed through a red filter and to be printed blue-green is evenly exposed in its yellow scale the third in C ink, yellow. On being Whirled the color sensations merge into one and should appear a neutral grey, as in Figure lb. If any color predominates it is necessary to reduce its.

ranging up from pur'e'white of the paper or support to the full intense saturation of the standardized ink, as balanced on the test dish. The grey scale is printed to equal that gradation of grey which would result from the mixture of tne primary color scales, and willbe degraded from pure white to deep black.

The preferred method of preparing the color tint strips is by printing each from a standard plate but successively diluting the ink with tint base or varnish, which may be according to the following schedule.

Parts of ad- Parts of Step justed ink tint base Percent Percent 1 100 0. 2"" s3 11 3"" 66. 4 3a. 4"" 49.8 50. 5 33. 2 66. 6": 16.6 as. 7 purest white paper 0.0 0.

A subject which is to be reproduced is arranged .as shown in Figure 3 with a control step strip I I the reproduction as it should be. However, the

red scale B, which should be of uniform and high density so that no blue green A pigment would be deposited in this area, does show gradation-having less opacity in the portions corresponding to the greater density of the original scale B. This is due to the fault of the B pigment in absorbing red light, resulting in less exposure of the negative from the darker areas of the B scale. This gradation would cause blue green ink to be printed which is obviously nonessential to the reproduction of a pure magenta, B scale.

Similarly in the green taken negative I313, the yellow C scale is evenly opaque and the grey and the magenta B scale show gradation, but a fault is apparent in the'bluescale A which'also shows gradation and would permit magenta ink to be printed in a scale which needs only bluegreen, A, ink for its perfect reproduction.

The blue taken, yellow printer negative [3C has the expected gradation in the grey and in the yellow C scale, but exhibits faults in both the redjB, scale and the blue, A, scale due to undesired absorption of the blue light by the blue-green, HA, and magenta, HB, step scales which should reflect blue light uniformly. This would result in yellow, C, ink being printed in all of the scales, while it should be printed only on the yellow and grey scales as reproduced. The grey printer I3C exhibits some gradation in all scales, which would result in degradation of these colors unnecessarily if uncompensated.

It may be noted that the above faults which I propose to correct depend upon the density of the original scale; consequently the means of correction should depend upon the density of the original scale. It will be seen for example that the faulty transparency of the C negative due to presence of blue green A ink in the subject may be corrected or neutralized by adding a density corresponding in some degree to the amount of blue green A ink in the subject.

By my procedure the measure of the density of the original scale is found in that separation negative properly showing its gradation. That is, the gradation of the A scale is shown in the A separation negative, of the B scale in the B negative, and of the C scale in the C negative.

To correct the C negative, for example, so that yellow dye will not be deposited in the bluegreen, A, scale, a positive transparency ISAC is made from the A negative of such contrast that the transparency of the A scale of the C negative is exactly corrected by the added density of the corrective positive, which will have the requisite characteristic since it is responsive to the color density of the original step scale.

The required degree of contrast is determined thru quantitative measurement of the density values of the various control scales, as is shown in the following tabulation of actual test values. A densitometer 14, as shown in Figure 5, being applied to' a negative I3, is customarily'employed for this purpose.

Sensitometric schedule Actual test data in process of correction of negatives for polychromatic reproduction. Figures tabulated are the densities at respective ends of the control step strips, and corresponding steps in the respective negatives and superposed masks.

Blue" Red Yellow Grey scale scale scale scale A B C G Density of original scales 140- 110- 0 90- 0 100- 0 Uncorrected b1ue" printer,

16-150 138-150 150-150 19-150 Corrective mask from red,"

l5 0- 0 12- 0 0- 0 28- 0 Total superposed densities 16-150 150-150 150 -150 47-150 Density difference, or range of corrected negative, 16A.-. 134 0 0 103 Uncorrected red" printer,

13 88-154 32-154 148-154 12-154 Mask from blue, 15 KB"... 76- 24- 10 10- 10 44- 10 Total superposed densities 164-164 56-164 164-164 56-164 Density difference of corrected negative, 163 0 108 0 108 Uncorrected yellow printer,

13G 71-103 58103 18-108 21-103 Mask from blue, AC 3&- 6 6- 6 6- 6 26- 6 Mask from "red," 1513C 0- 0 45- 0 0- 0 50- 0 Total superposed density. 109-109 1011-109 24-109 97-109 Density difierence, of corrected negative, 160 0 I 0 85 12 Uncorrected grey printer, 13G 76-173 118-173 173-173 21-173 Mask from yellow, l5CG 24- 8 38- 8 68- 8 62- 8 Total superposed densities. 100-181 156-181 241-181 83-181 Density difference of corrected negative, 16G 81 25 "-60" 98 deposit increases exponentially with the exposure, the factor of increase depending upon the degree of development,

0=E or more simply and usably, log 0: log E over a considerable range of exposure, exposure being a product function of light intensity and the duration of its action.

same emulsion: and same degree of develop- For convenience, log 0 is termed Density, D and now for two different exposures on the ment, 7: D1=7 lOgEr and Dz='y log E2 Subtracting, D1D2='y (10g E1'10g E'2) and i s I D D, Density difference 10g Ej-log E -log exposure difference Now in printing, the log exposure difference is equivalent to the available density difference of the negative being printed, so that to obtain a desired density diiference or range in a positive from printing an available density difference in a negative the development contrast required will be:'

Density difference required in positive Density difference available in negative For practical attainment of this calculated value it is advisable to standardize development conditions and prepare a chart similar to Figure 4, which shows the gamma obtained at a given duration of development. Having found the required gamma, the time of development necessary is found from this graph.

As an example of the calculations and corrections applied to correct any deficiency the correction of the C or yellow printer is explained in detail. First observing the A scale of the nega-- tive I3C, it is noted that a density difference of 32 must be corrected by an inverse or positive density difference of 32. This must be responsive to the presence of blue green in the original, which is measured by the gradation of the A scale in the negative 13A. Here the density difference is 134 which must be so printed that a; positive of density difference 32 results.

and such time of development, three and one half minutes as shown by the'chart, for this contrast is given a properly exposed print from thenegative "A. This positive I'5AC is superposed accurately on the negative I30.

Now proceeding to correct the faulty vB scale of the negative it is seen that a density difference of 45 must becompensated. The B scale 'of the I B negative which shall control this correction in accord with the presence of magenta, is seen to have an available density difference of 122.

The positive |5BC is developed to this contrast, six minutes according to Figure 4, and is fastened accurately in superposition upon negative I30 and the previous corrective mask ISAC, to form in combination the corrected separation negative IBC. A thin base film is used to advantage for these corrective masks, as it permits more precise registry.

n referring to the density measurement tabulation it will be seen that on addition of the.

superposed densities at respective ends of the B scale the combination density is equal at each extreme, no gradation remains, and no yellow, C,

ink will be placed on the magenta, B, scala' Similarly it is evident that there is no gradation in the A scale and this blue green scale will not graphically in Figure 3. The white wedge faults in the negative are precisely filled by the black Wedge corrective densities in the respective corrective masks when superposed in registry; and gradation is reduced in the grey strip portion G.

It is notable that a density difference of only 12 remains for gradation of yellow ink on the grey scale G while a density difference of about 100 is noted in the others. This may be understood on realizing that the blue-green A, and magenta B, inks, due to their faults, are absorbing blue light-which is properly the duty of yellow, C, ink. Therefore, some of its work being done, less yellow C, ink need be deposited, and this correction has been precisely effected, responsive to the presence of these inks.

By similar corrective procedure the A and B negatives have been corrected so that gradation is present only where ink should be deposited.

The uncorrected grey printer negative l-3G,

whatever the method employed to obtain it, is

subject to correction by superposition-of a posi tive mask ISCG printed from the uncorrected yellow printer negative I 3C. This results in brightening those portions of the reproduction where yellow is to be printed and to a lesser degree where magenta and blue-green are to be printed, these latter effects being due to use of l the uncorrected C negative in printing the mask. Use of my method of control, however, makes the use of the grey printer far less essential than hitherto. Previous use of a black printer has served-two functions-to define delicate details which might be indistinct thru non registry of the three color plates and secondly, to cover up and hide the faults which I eliminate by color responsive correction. These faults become increasingly obvious with greater depth of the superposed inks, that is, in the dark portions; printers correct the faulty hue by a heavy overcoating of black. My use of the grey printer is directed only to the former purpose, and preparation and masking of my grey printer is designed to' permit this delineation of detail without unnecessary degradation of the brighter colors. Use of a mask printed from the uncorrected C negative permits its faults in the A and B scales to be effectively utilized to this end.

densities of the other inks to be present in the reproduction. However, these further refinements and corrections are both economically impractical and artistically unnecessary; for with available inks of good quality there is practically no gradation in the C scale of A or B negatives. There remains the possibility that because of their phenomenal permanency one would wish to use pigments of poor color quality. In this case these faults could be compensated in all the tint step scales. However, the brightest white would be darkened to a grey and light tints of any color would be somewhat degraded-in fact, this is' the only reason for choosing the most pure inks available for use in my process.

Having now obtained properly corrected separation negatives there remains only to print from them in a standardized manner to obtain proper printingof the original scene.

The printing plates are so prepared from each separation negative that the color tint step scale to be printed by each will be reproduced in its original contrast and density. To achieve density balance in' the positives a test exposure is made from any corrected separationnegative, the exposure for the others being calculated logarithmically from their known density values. To obtain the required density range in the several color -tint step strips, gamma'and time of development are calculated as described for the masks. For example, if an exposure of 11 seconds were found suitable for the yellow negative having a maximum density of 109 in the C scale, the exposures for the red printer B negative having a density of 164 in the B scale and for the blue printer A negative having a maximum density of and the grey printer, density maximum would be 40 seconds, 29 seconds and 59 seconds respectively, and development be such as to give the desired density difference. From these balanced positives l1, mechanical print surfaces l8 of qualities equivalent to those used in preparation of the original color disc and color tint step strips are prepared.

It will be understood that having correctly reproduced the separate tint strips and grey, any color or combination of colors. in the original will be correctly synthesized in the reproduction.

While I have described my process as applied to three color subtractive reproduction it is evidently applicable to multi-color separation fault correction however complex by use of additional masks made according to my method, and I do not limit my claim to the specific embodiment herein set forth.

I claim:

In a process of color reproduction, preparation of a master stepped density strip, measuring density difference between end steps thereof, preparing therefrom printing plates by identical and standardized methods, inkingup and printing from said plates upon a support medium such as will be used in the final reproduction, forming a tint step strip in each of the ink pigments to be used as components constiany initial separation record should any reproduced strip therein have a density-difference, except that which is complementary m the sens-itivity used in making the initial separation record, said mask being photographically printed from and being of Opposite sense of tonal gradation to, that initial separation record made equal to the ratio: density difference in original with sensitivity complementary to the original pigment strip which has a density difference requiring masking in the initial separation record to be masked,- said mask being developed to a contrast equal to the ratio: density difference in said strip in the initial separation record to be masked density difference in corresponding strip in the negative from which mask is printed, securing said mask in registry with the initial separation record for which designed, to form corrected initial separation records, photograph ically printing corrected secondary separation,

records therefrom, being developed toa contrast, 20

' initial separation record, and printing said plates master stepped density strip density diiierence in corrected initial separation record reproduc tion of step strip complementary to the sensitivity used in making the initial separation rec- 0rd, preparing printing plates from these corrected secondary records by the identical standardized methods usedin preparing the plates for the original tint step control strips, adjusting exposures so that the end steps of the reproduced tint strips are of value equal to that of their original printing plates, inkingup each plate in that original ink whose pigment is complementary to the sensitivity used in making the successively in superposition upon a support medium such as was used in printing the control I strips.

SALVADOR GARCIA GUERRERO.

Images