GB2240001A

GB2240001A - Film characteristic simulator

Info

Publication number: GB2240001A
Application number: GB9000889A
Authority: GB
Inventors: Terrence William Mead; Delemos Richard Brooks
Original assignee: Rank Cintel Ltd
Current assignee: Rank Cintel Ltd
Priority date: 1990-01-16
Filing date: 1990-01-16
Publication date: 1991-07-17
Anticipated expiration: 2010-01-16
Also published as: GB2240001B; GB9000889D0; WO1991011071A1

Abstract

To simulate non-linear transfer characteristics of photographic film to video a digital gamma correction circuit is provided with additional circuitry to vary the gamma correction factor Y applied to the logarithm of the input video X. Black level onset and magnitude values are selected (CNTL 3, CNTL 2) and applied to the linear gamma correction level (CNTL 1). Similar values are selected for the white level. Both sets of values and the linear value are then applied to multiplier (12) of the gamma correction circuit as a non-linear gamma correction factor.

Description

FILM CHARACTERISTIC SIMULATOR This invention relates to video special effects and in particular to the control of film transfer characteristics in telecine or film writing apparatus.

In telecine and other television studio equipment it is necessary to correct for the non-linear transfer law of the receiving Cathode Ray Tube (C.R.T.) and for other non-linearities that occur, for example, in the film. This correction is often performed by a circuit which multiplies the log of the input signal and then takes its anti-log. Such a circuit is known as a gamma correction circuit, a typical example of which is shown in Figure 1. In that circuit the log of a digital input video signal X is taken at 10 and that signal log X is multiplied at 12 by a gamma control input Y to give an output signal Y log X. The anti-log of this signal is taken at 14 resulting in an output XY where Y is the gamma of the output. Conveniently, both the log and anti-log units 10 and 14 are look-up tables.

The present invention is defined by the claims to which reference should be made.

In a preferred embodiment of the invention an incremental gamma level adjustment is derived for the regions of the video output range approaching black and white peak values. The onset point of the adjustments and the size and polarity of the increments may all be selected. Thus, the effects which could previously only be obtained by operating the film in the non-linear parts may, using the apparatus and method embodying the invention be obtained electronically using, for example, a telecine, and film exposed in the linear part of its transfer characteristic. Furthermore, the same effect may be produced with video signals produced, for example, from live cameras during editing. Thus, the invention allows film effects to be introduced into video signals not produced from film.

An embodiment of the invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a block diagram of the prior art gamna correction circuit previously referred to; Figure 2 is a graph of film density against the log of the output video level for a telecine; and Figure 3 is a block diagram of apparatus embodying the invention.

The transfer law of photographic film typically demonstrates an S shaped characteristic as illustrated by plot 20 in figure 2. As the uppermost and lpwermost portions of the characteristic curve are not linear, film is presently operated in the linear region. Where special effects or ircods are required, film producers will often adjust lighting and exposure levels to drive the film into the non-linear extremities. However, production of such effects is difficult to achieve and expensive as it often involves reshooting film.

Further, it introduces time delays into film production.

The circuit of figure 3 enables these artistic effects to be reproduced electronically, so avoiding the need to reshoot film in the extreme regions of the curve. Furthermore, it enables effects hitherto only available with photographic film to be available for electronic editing of material from live television cameras and other instances where the effect is created by adjustment of a video signal.

In the circuit of figure 3, the prior art digital gamma correction circuit is provided with additional circuitry to enable the gamma input of multiplier 12 to be varied in a number of different ways according to the effect desired. The effects are controlled by a number of different control signals CML 1 to CNTL 5 the functions of which and the areas of the transfer curve on which they operate are shown in figure 2.The elements controlled by each control signal are as follows: CML 1: Control overall gamma value CML 2: Control amplitude and polarity of black gamma CML 3: Control onset of black gamma CML 4: Control amplitude and polarity of white gamma CML 5: Control onset of white gamma The circuit comprises indentical sections, one of which adjusts the white section of the S curve and the other the black section.

The black section adjustment comprises an adder 22 having as its one input log X and as its other input CML 3, multiplier 24 which multiplies the clipped output of adder 22 with CML 3 and a further adder 26 which adds CML 1 to the output of the multiplier 24. .The white section adjustment comprises adders 28 and 32 and multiplier 30 arranged as for the black section adjustment except that rather than having a separate control input, the final adder 32 adds the output of the black adjustment to the output of multiplier 30. The output of adder 32 forms the gamna input Y of multiplier 12 of the digital gamma correction circuit of figure 1.

To understand how CML 1 acts as the overall gamma control over the whole range, consider how the circuit operates when control inputs 2 and 4 (adjustable black and white gamma) are set to zero. The output from multipliers 24 and 30 will be zero and the output of white adder 32 will be CML 1. Thus, Y = COIL 1.

Consider now how the black and white gamma levels can be added to the overall gamma control. Adder 28 substracts an adjustable onset from the digital video input (log X) and the following clip circuit 29 removes any resulting negative numbers with the result that a correcting signal is generated when the signal is near white. The correcting signal is proportional to the video so that its onset is gradual. This signal is supplied as an input to multiplier 30 where it is multiplied with CML 4. CML 4 is bipolar determines both the severity and the polarity of the increase in white gamma. The output of multiplier 30 is added by adder 32 to the overall gamma control which is the output of adder 26 (assuming for the moment that CML = O).The result of CML 5 and 4 has no effect over most of the signal range, but as the video output level rises to the adjustable onset level (determined by the value of CML 5) the gamma valve is increasingly modified up to the peak, white, level. The severity or polarity of the modification is controlled by CML 4.

Similarly to the above, the black gamma control subtracts an ajustable onset from the video input (log X). This function is provided by adder 22, the adjustable black onset being CML 3. The output from adder 22 is clipped by clip circuit 23 to remove any negative values and the resultant applied as an input to multiplier 24 where it is multiplied by the adjustable black gamma control CML 2. Again, the presence of the clip circuit 23 means that corrections will only be made over a small part of the circuit (when the output of adder 22 is positive). In practice, a correcting signal is generated whenever the signal is near black.The correcting signal again is proportional to the input video so that its onset is gradual CML 2 is a bipolar signal and the output of multiplier 24 is added to the overall gamma value CML 1 in adder 26. Control inputs 2 and 3 have no effect over most of the signal range but modified increasingly the incremental gamma value down to the minimum, black, level from a point determined by the black level onset value CML 3. The polarity and severity of the effect can be adjusted by adjusting CML 2.

Thus the circuit of figure 3 enables a video signal to be adjusted to simulate electronically effects that could otherwise only be obtained by operating photographic film in the non-linear parts of the transfer characteristic. The circuit has the advantage that all the control values CML 1 to 5 may adjustable so that any desired film characteristic may be simulated. Furthermore, the circuit is not limited to the post-processing of film in a telecine to create effects that could havew been generated during shooting of the film.

The same effects may be applied to a video signal that has not been derived from photographic film but, for example, from a live television camera, or to compensate for a film which has been incorrectly exposed.

Claims

1. A method of adjusting a video signal to simulate a pre-determined non-linear film transfer characteristic, comprising selecting a gamma correction value to be applied to the video signal across the signal range, deriving an adjustment to the selected gamma correction value for at least a portion of the signal range, modifying the selected correction value to include the derived adjustment, and applying the modified gamma correction value to the input video.

2. A method according to claim 1, wherein the portion of the signal range for which an adjustment to the selected gamma correction value is determined is a region approaching an extremity of the signal range.

3. A method according to claim 2, wherein the onset point of the modification to include the adjustment is variable.

4. A method according to claim 3, wherein the degree of adjustment is incremental from the onset value to the extreme value of the signal.

5 A method according to claim 4, wherein the size and polarity of the incremental adjustments of the gamma adjustment are variable.

6. A method according to any of claims 2 to 5, wherein an adjustment is selected for two regions of the signal range, a first region approaching the black signal extremity and a second region approaching the white signal extremity.

7. A method according to any of claims 1 to 6, wherein the output video signal is the anti-logarithm of the product of the liodified gamma correction value and the logarithm of the input video signal.

8. A method according to any preceding claim, wherein the pre-determined non-linear film transfer characteristic is chosen to simulate transfer of photographic film to video.

9. Apparatus for adjusting a video signal to simulate a pre-determined non-linear film transfer characteristic, comprising means for providing a linear gamma correction value across the video signal range, means for deriving, for a selected portion of the video signal range, an adjustment to the selected gamma correction value, means for modifying the linear gamma correction value to include the adjustment, and means for applying the modified gamma correction value to the input video signal.

10. Apparatus according to claim 9, wherein the adjustment derivation means comprises means for deriving an adjustment for the portion of the signal range approaching a first extremity of the signal range.

11. Appartus according to claim 10, comprising means for deriving a further adjustment for a portion of the signal range approaching the opposite extremity of the signal range of the first extremity.

12. Apparatus according to claim 10 or 11, wherein the modifying means comprises means for selecting the onset point on the video signal range at which the derived adjustment is included with the linear gamma correction value.

13. Apparatus according to claim 12, including means for incrementing the adjustment value from the onset point to the extreme value.

14. Apparatus according to claim 13, comprising means for varying the size and polarity of the incremented adjustment values.

15. Apparatus according to any of claims 10 to 14, wherein the means for selecting the onset point on the video signal range from the peak value at which the linear gamma correction value is modified comprises means for subtracting an adjustable control value from the input video signal means for removing negative value of the subtraction, and means for multiplying the remaining positive values by a scaling factor indicative of the degree of adjustment.

16. A method of adjusting a video signal to simulate a pre-determined non- linear film transfer characteristic substantially as herein described with reference to figures 2 and 3 of the accompanying drawings.

17. Apparatus for adjusting a video signal to simulate a pre-determined non-linear film transfer characteristic substantially as herein described with reference to figures 2 and 3 of the accompanying drawings.