GB1579738A - Colour picture reproduction system - Google Patents

Description

(54) COLOUR PICTURE REPRODUCTION SYSTEM (71) We, KODAK LIMITED, a Company registered under the law of England, of Kodak House, Station Road, Hemel Hempstead, Hertfordshire, do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement; This invention relates to the production of video signals.

Video signals representative of a picture, or scene, are produced, in the case of a television camera, by electronically scanning an image or images of the picture, or scene, and processing the output signal(s) to the required form. Where a colour video signal is required it is known to apply the output signals to a matrix unit to obtain colour masking.

One difficulty encountered with electronic scanning systems, in television cameras and elsewhere, is that the signal/noise ratio is low when the picture, or scene, is badly lit or when the picture is very dense, and this can lead to degraded colour reproduction tor such areas when the video signal is used to form a display. An object of the present invention is to provide some compensation where such colour degradation would otherwise occur.

In accordance with the present invention there is provided electronic scanning apparatus adapted to produce a colour video signal output, the apparatus having a colour masking matrix arranged to receive and modify the colour signals produced by scanning a picture; wherein switch means responsive to the signals indicating a predetermined low light level or high optical density or high colour saturation at a scanned picture point are arranged to switch an auxiliary matrix into operation, this auxiliary matrix being arranged to receive the corresponding output signals from the colour masking matrix and to modify these signals so that the colour video signal output includes improved colour information for these picture points.

The apparatus may include more than one auxiliary matrix, the switch means being arranged to switch the auxiliary matrices selectively into operation in dependence upon the magnitude of signals corresponding to the scanned picture point. The switch means may switch the auxiliary matrices into and out of operation in a progressive manner.

The apparatus may include means for dividing each output signal from the colour masking matrix into upper and lower magnitude parts, and applying only one of these parts to an auxiliary matrix.

A first auxiliary matrix may be a unity matrix, and a second auxiliary matrix may be a colour compensating matrix.

A first auxiliary matrix may provide weak matrixing for at least one of the primary colours for which the scanning signals are derived, and a second auxiliary matrix may provide desaturating compensation.

The switch means may operate in response to the light level, optical density or colour saturation in all of red, green and blue signals from the colour masking matrix all of the colour signals being checked independently.

Alternatively the switch means may operate in response to a signal derived from an average of the lower magnitude parts of the output signals from the colour masking matrix.

According to another aspect of the invention, an auxiliary matrix is arranged to operate on an extreme range of one or more of the modified colour signals, the extreme range being one in which a colour signal has a non-linear image representation characteristic, to produce a changed set of video signals that are driven relatively towards neutral, and switch means are arranged to control the contribution of the changed set of video signals to the colour video signals output in accordance with the optical density at the scanned picture points.

The invention will normally be used with television cameras and film scanners of the conventional kind in which red, green and blue density signals of the scanned picture point are available, but it is not limited to such colours or devices.

The invention may be applied to an electronic film scanner in apparatus for producing a photographic print from a transparency, and two such applications will now be described by way of example with reference to the drawing accompanying the Provisional Specification in which Figure 1 is a block diagram of the relevant portion of the second example of signal processing circuit for red, green and blue signals; and in which Figure 2 is a block diagram of a fourth example of signal processing circuit, but for simplicity showing only the circuitry for the red signal.

Referring now to Figure 1, which is identical for the first and second embodiments except in so far as matrix unit 14 is concerned, there are shown three photomultipliers 1, 2 and 3 respectively sensitive to red, green and blue light reflected by mirrors 5 as it leaves the transparency 4 during scanning at the usual television scanning rate by a flying spot scanner having a cathode ray tube. The resultant red, green and blue transmission signals are applied to respective logarithmic amplifiers 7, 8 and 9 and the outputs thereof pass to a conventional electronic colour mask 10. The matrixed signals therefrom are applied to a divider or limiting amplifiers, 11/12 such that signals corresponding to a scanned picture point of a transparency having an optical density of less than a predetermined value appear in divider portion 11 and are applied to a summing amplifier 13 for the colour video output signal. Matrixed signals corresponding to picture points whose transparency density is more than the predetermined value appear in divider portion 12 and are applied to auxiliary matrix units 14 and 15, and the output from these auxiliary matrix units is also applied to the summing amplifier 13, the composition of these outputs being controlled by a selector switch 16 which is responsive to a control signal derived by a control signal generator 17.

The control signal gencrator operates when the transparency density of the scanned picture point reaches a selected value.

The predetermined value referred to above in connection with the divider 11/12 has been chosen as 2.4 in carrying the invention into effect since this is the greatest optical density expected from a normally exposed transparency or slide. A different value could be chosen, if necessary, where the invention is applied to a television camera.

In this embodiment the second auxiliary matrix unit is used to modify the colour video output signal in such a way as to neutralise shadow areas of a point made from the transparency using the colour video output signal to reconstitute the picture in known manner and this compensation is effected without degrading saturated colours. For this purpose the matrix unit 15 with all positive terms combines the upper parts of the density signals which are applied thereto, so as to make the picture point more neutral. Where the scanned point is a saturated colour then desaturation of the picture point is not desired and the required output will be derived from matrix unit 14 which is a unity matrix in this first embodiment. In other words the matrix terms for unit 14 should be 1 in this first embodiment, and not as shown in Figure 1.

It appears that a suitable instant for initiating switching in this arrangement, using selector switch 16, is when the red, green and blue densities rise together above 1.8. This point may be determined by means of a circuit which can make the three comparisons necessary independently and which can then actuate the selector switch 16 at the appropriate time. In practice however it is preferred to use a simplified circuit that produces a control signal which is the average of the lower parts of the red, green and blue density signals, that is, the average of the output of the divider 11. Switching then takes place at a density of 2.2, although in practice, to avoid contouring effects it is preferred that the switching between the matrix units 14 and 15 is progressive, beginning at a density of 2.0 and being complete at a density of 2.4. Thus if the scanned picture point is a red having densities R < 1.8, G > 2.4, B > 2.4 then the control signal is less than 2.2, the unity matrix 14 remains operative and the saturation is maintained. A near-neutral picture point with R > 2.4, G > 2.4, B > 2.4 will however cause the desaturating matrix unit 15 to be switched in, and the colour of the resultant picture will be made more nearly neutral by the matrixing operation in unit 15.

The specific system may be summarised in matrix terms as follows where R', G', and B' are the red, green and blue output signals from the summer 13; AR, AG and AB are the lower parts of the divided density signals (from divider 11); and AR', AG' and AB' are the upper parts of the divided density signals (from divider 12).

A simplification of this specific system, resulting in what will be called the second embodiment, is possible if a compromise term value of .75 is adopted for the diagonal terms of both the matrix 14 and the desaturating matrix 15. Switching of the off-diagonal terms only of these matrices is then necessary in changing over from one matrix unit to the other.

In this case the matrix 14 is not a unity matrix, and the matrix terms shown in Figure 1 apply.

Improved colour reproduction in prints made from transparencies using a colour video output signal modified by the above techniques has been noted.

The embodiments described above can be changed to modify the colour video output signal so as to add detail in saturated coloured area of a print from a transparency. This is effective for example where there is some red information in the transparency in the form of modulation of the small amount of cyan dye, and where this corresponds to a point off the toe of the characteristic curve of the printing paper emulsion. This may be compensated for in the third embodiment by introducing some weak matrixing into the red channel of matrix 14 of Figure 1. Thus the same general arrangement is adopted as above but the matrices are modified as follows:-

The value of .06 was found by experiment to give an acceptable amount of modelling to reds (with G = B > 2.4) without adding too much 'wrong' information from the green and blue channels, and without causing yellows (with B > 2.4) to be reproduced too green.

Dark neutrals are not affected by the imbalance of the row sums in the first matrix, because they are handled by the second matrix 15.

A fourth embodiment will now be described with reference to Figure 2 of the drawings.

Bearing in mind that this block diagram is concerned mainly with the red channel of the signal processing circuit it will be seen that it corresponds broadly to the second embodiment except in the derivation of the control signal for effecting switching, and that the control signal differs from channel to channel.

As before, red, green and blue transmission signals are obtained from photomultipliers and logarithmic amplifiers (7, 8 and 9 in Figure 1) and applied to a conventional electronic colour mask 20. The matrixed signals therefrom are applied to a divider 21/22 which separates the red signal R, for example, into AR and AR', where AR is the signal corresponding to an optical density D between 0 and 2.4 of the scanned picture point of the transparency, and AR' corresponds to D above 2.4. Similarly the green signal G = AG + AG', and the blue signal B = AB + AB', these signals being divided at the same optical densities.

From the divider 22 the signal AR' is applied to the first auxiliary matrix unit 23 which in this case is not a unity matrix but a 0.75 multiplier. The output from matrix unit 23 is fed to a summing amplifier 24 together with the output AR of divider 21.

From the inputs and output of divider 21 the signals, for the red channel, AR + B + G are applied to the summing amplifier 25 whose output passes to divide-by-3 divider 26 and thence to a level dependent switch 27. Switch 27 operates when its input, AR + G + B 3 is greater than 2.4 and allows the excess over 2.4 to pass. For a neutral colour where R = G = B, each greater than 2.4, then the output from switch 27 is AG' + AB' 3 Now the required output from the red channel for a neutral input in embodiment 2 is shown by R' = AR + 0.75AR' + 0.25AG' + 0.25AB', and the output from switch 27 is therefore weighted by 0.25 x 3 to give the required coefficients for AG' and AB'. The 0.75 multiplier 28 produces as output, the positive part of 0.75 (AR .3G. S - 2.4) and applies it to the summing amplifier 24.

Thus for a neutral colour, R = G = B > 2.4, the output of summing amplifier 24 is R'AR+o.7sARl+o.sH(AR+3+B - 2.4) 3 where H is the positive part.

For lighter or more saturated colours, the signal AR + G + B < 2.4 3 and there will be no output from switch 27. The output from summing amplifier 24 is then R' = AR + 0.75 R'.

In this fourth embodiment the desaturation or neutralisation is effected more gradually.

For example where the scanned picture spot is a red, G = B = 3, desaturation commences when the red density excceds 1.2 but does not reach its maximum until D is greater than 2.4.

The system may be represented in matrix form as follows:

The fourth embodiment may be modified to provide some modelling, i.e. addition of details in a saturated colour point whose optical density exceeds 2.4, as in the third embodiment. A suitable matrix for this fifth embodiment is as follows:

In the above embodiments there are two auxiliary matrix units, one possibly being a unity matrix. The invention may alternatively include a larger number of matrices, particularly where reversal films with large non-linear interimage effects are involved. The appropriate matrix would then be switched in according to the density or saturation of each scanned picture point in the film.

The preceding embodiments have been concerned with film scanners but it will be appreciated that the invention is applicable to television cameras. Thus the first embodiment can be applied to such a camera so that in the subsequent television display the shadow areas are reproduced nearer neutral and yet saturated colours are unaffected by this compensation. Further, since television systems have a sharp toe to their transfer characteristics, saturated colours are often reproduced with little detail in them. The addition of weak matrixing in all of the three channels, using the technique adopted for the third and fourth embodiments will then compensate for this effect.

As indicated above, the invention is useful where there is a high optical density at a scanned picture point. It should be appreciated that the undesirable effects which can be compensated for by use of the invention, where the film or slide information is at high density, include shading errors, mismatches in electronic circuits such as logarithmic amplifiers, and mismatch in the scanned film.

WHAT WE CLAIM IS: 1. Electronic scanning apparatus adapted to produce a colour video signal output, the apparatus having a colour masking matrix arranged to receive and modify the colour signals produced by scanning a picture; wherein switch means responsive to the signals indicating a predetermined low light level or high optical density or high colour saturation at a scanned picture point are arranged to switch an auxiliary matrix into operation, this auxiliary matrix being arranged to receive the corresponding output signals from the colour masking matrix and to modify these signals so that the colour video signal output includes improved colour information for these picture points.

2. Apparatus as claimed in Claim 1, including more than one auxiliary matrix, the switch means being arranged to switch the auxiliary matrices selectively into operation in dependence upon the magnitude of the signals corresponding to the scanned picture point.

3. Apparatus as claimed in Claim 2, wherein the switch means is arranged to switch the auxiliary matrices into and out of operation in a progressive manner.

4. Apparatus as claimed in Claim 1, including means for dividing each output signal from the colour masking matrix into upper and lower magnitude parts, and applying only one of these parts to an auxiliary matrix.

5. Apparatus as claimed in Claim 2 or Claim 3, wherein a first auxiliary matrix is a unity matrix, and a second auxiliary matrix is a colour compensating matrix.

6. Apparatus as claimed in Claim 2 or Claim 3, wherein a first auxiliary matrix provides weak matrixing for at least one of the primary colours for which the scanning signals are derived, and a second auxiliary matrix provides desaturating compensation.

7. Apparatus as claimed in any preceding claim, wherein the switch means operates in response to the light level, optical density or colour saturation in all of red, green and blue signals being checked independently.

8. Apparatus as claimed in Claim 4, wherein the switch means operates in response to a signal derived from an average of the lower magnitude parts of the output signal from the colour masking matrix.

9. A television camera including apparatus as claimed in any preceding claim.

10. A photographic film scanner including apparatus as claimed in any of Claims 1 to 8.

11. Electronic scanning apparatus substantially as hereinbefore described with reference to Figure 1 or to Figure 2 of the drawings accompanying the Provisional Specification.

12. Electronic scanning apparatus adapted to produce a colour video signal output, the apparatus having a colour masking matrix arranged to receive and modify the colour signals produced by scanning a picture; wherein an auxiliary matrix is arranged to operate on an extreme range of one or more of the modified colour signals, the extreme range being one in which a colour signal has a non-linear image representation characteristic, to produce a changed set of video signals that are driven relatively towards neutral, and switch means are arranged to control the contribution of the changed set of video signals to the colour video signal output in accordance with the optical density at the scanned picture points.

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (12)

**WARNING** start of CLMS field may overlap end of DESC **. In the above embodiments there are two auxiliary matrix units, one possibly being a unity matrix. The invention may alternatively include a larger number of matrices, particularly where reversal films with large non-linear interimage effects are involved. The appropriate matrix would then be switched in according to the density or saturation of each scanned picture point in the film. The preceding embodiments have been concerned with film scanners but it will be appreciated that the invention is applicable to television cameras. Thus the first embodiment can be applied to such a camera so that in the subsequent television display the shadow areas are reproduced nearer neutral and yet saturated colours are unaffected by this compensation. Further, since television systems have a sharp toe to their transfer characteristics, saturated colours are often reproduced with little detail in them. The addition of weak matrixing in all of the three channels, using the technique adopted for the third and fourth embodiments will then compensate for this effect. As indicated above, the invention is useful where there is a high optical density at a scanned picture point. It should be appreciated that the undesirable effects which can be compensated for by use of the invention, where the film or slide information is at high density, include shading errors, mismatches in electronic circuits such as logarithmic amplifiers, and mismatch in the scanned film. WHAT WE CLAIM IS:

1. Electronic scanning apparatus adapted to produce a colour video signal output, the apparatus having a colour masking matrix arranged to receive and modify the colour signals produced by scanning a picture; wherein switch means responsive to the signals indicating a predetermined low light level or high optical density or high colour saturation at a scanned picture point are arranged to switch an auxiliary matrix into operation, this auxiliary matrix being arranged to receive the corresponding output signals from the colour masking matrix and to modify these signals so that the colour video signal output includes improved colour information for these picture points.

2. Apparatus as claimed in Claim 1, including more than one auxiliary matrix, the switch means being arranged to switch the auxiliary matrices selectively into operation in dependence upon the magnitude of the signals corresponding to the scanned picture point.

3. Apparatus as claimed in Claim 2, wherein the switch means is arranged to switch the auxiliary matrices into and out of operation in a progressive manner.

4. Apparatus as claimed in Claim 1, including means for dividing each output signal from the colour masking matrix into upper and lower magnitude parts, and applying only one of these parts to an auxiliary matrix.

5. Apparatus as claimed in Claim 2 or Claim 3, wherein a first auxiliary matrix is a unity matrix, and a second auxiliary matrix is a colour compensating matrix.

6. Apparatus as claimed in Claim 2 or Claim 3, wherein a first auxiliary matrix provides weak matrixing for at least one of the primary colours for which the scanning signals are derived, and a second auxiliary matrix provides desaturating compensation.

7. Apparatus as claimed in any preceding claim, wherein the switch means operates in response to the light level, optical density or colour saturation in all of red, green and blue signals being checked independently.

8. Apparatus as claimed in Claim 4, wherein the switch means operates in response to a signal derived from an average of the lower magnitude parts of the output signal from the colour masking matrix.

9. A television camera including apparatus as claimed in any preceding claim.

10. A photographic film scanner including apparatus as claimed in any of Claims 1 to 8.

11. Electronic scanning apparatus substantially as hereinbefore described with reference to Figure 1 or to Figure 2 of the drawings accompanying the Provisional Specification.

12. Electronic scanning apparatus adapted to produce a colour video signal output, the apparatus having a colour masking matrix arranged to receive and modify the colour signals produced by scanning a picture; wherein an auxiliary matrix is arranged to operate on an extreme range of one or more of the modified colour signals, the extreme range being one in which a colour signal has a non-linear image representation characteristic, to produce a changed set of video signals that are driven relatively towards neutral, and switch means are arranged to control the contribution of the changed set of video signals to the colour video signal output in accordance with the optical density at the scanned picture points.