GB2325369A - Stabilisation of photoelectric cells - Google Patents

Abstract

Photoelectric cell drift, for example in a flying spot scanner, is minimised by arranging a stable independent light source such as an LED (110) adjacent the cell window (30) and illuminating the window with this light source at a time when the primary light source is not incident the cell, for example during a frame blanking interval. The output of the cell is compared (60) to a reference and the cell gain corrected if necessary. The reference is a software generated 'virtual' target value based on the transfer characteristic of the cell.

Description

1 2325369 STABILISATION OF PHOTOELECTRIC CELLS This invention relates to

the stabilisation of photoelectric cells and the avoidance of output drift over time. It is particularly, but not exclusively, related to drift in photomultiplier tubes used in flying spot film scanners.

Figure 1 illustrates schematically a basic flying spot film scanner 10. Film 12 passing through a film gate 14 is scanned by a raster generated by a flying spot scanner 16. The raster beam is modulated according to the film density at each point of the film frame being scanned at a given time. The modulated light is detected by a photomultiplier tube 18 which produces an output voltage which varies with film density. In practice, three photomultiplier tubes are used, responsive respectively to Red, Green and Blue wavelengths. The three colour signals produced are then processed in parallel. For ease of explanation, only a single channel is shown in Figure 1. The output of the photomultiplier 18 may then be corrected for afterglow from the cathode ray tube phosphor and then is converted to a digital signal 20 in most modern scanners. The signal may then be subject to a range of processing such as shading correction, colour processing and correction and aperture correction.

The resultant processed signal is then incorporated into a chosen video signal format such as R G B or YC,C, for broadcast, manipulation or store.

A film scanner such as has been described may be found, for example, as part of telecines made and sold by Cintel International Limited of Hertfordshire, England under the trade marks URSA and URSA GOLD.

The photomultiplier tubes comprise a photocathode which emits electrons in response to incident light photons; and an anode consisting of a number, typically ten, dynodes. The dynode multiplier operates by secondary emission whereby each electron arriving at the first dynode results in, say, two electrons being directed to the next dynode, where a similar gain takes place. The gain of the photomultiplier tube is controlled by varying the voltage applied to each stage of the dynode multiplier.

A telecine has to operate using either positive print film or original camera negative film. With positive print film image 'blacks' correspond to low values of CRT light reaching the photoelectric cell in the form of a photomultiplier, and conversely high values of CRT light reach the photoelectric cell with image 'whites'. With negative film the position is reversed with 'blacks' producing high values of CRT light (because the negative image is clear) and 'whites' correspond to low values.

The current signal generated in flying spot telecine is generated in a photomultiplier and is between 50 and 100pA. With good design the head amplifier noise may be considered negligible, the dominant noise is the shot noise which is a function of the number of photons striking the photosensitive cathode and the number of electrons released in the photomultiplier. The noise is therefore a function of the incident light and is proportional to the square root of the signal current, falling to zero at black level (i.e. with no PMT signal current).

Any electrical gain applied in a telecine processing channel will also add noise with increasing gain producing increasing noise. Thus, the dominant noise for any given film density and film type will approximately vary according to the following table:

- 3 FILM TYPE IMAGE WHITE IMAGE BLACK Shot Electrical Shot Electrical Noise Noise Noise Noise Print High Low Low High Negative Low High High Clearly there is therefore a requirement for the telecine operator to be able to optimise the signal to noise ratio by varying the photoelectric cell gain., and electrical gain dependent upon the type of film used, and the particular density ranges of the film.

Figure 2 shows the current gain: anode voltage transfer characteristic of photomultiplier tubes (PMTs). This is logarithmic. However, PMTs are prone to drift with time. This is due to a number of effects including thermionic emission of electrons by the photocathode; ionisation of residual gases; glass scintillation; ohmic leakage current; field emission (electron extraction); accumulated light exposure in the short to medium term; and by damage caused to the last dynode of the dynode multiplier by heavy electron bombardment. A typical set of drift characteristics is shown in Figure 3. It will be seen that there may be as much as 10% drift over one hour and 15-20% over 10 hours. Such high drifts have been measured on telecines operated by Cintel International Limited. As telecines operate for many hours at a time it is very important to be able to control these drifts and a number of techniques are known in the art.

In our prior publication GB-A-2215551, the dri'ft in gain of the PMT is controlled by generating a reference value and periodically during the frame blanking intervals, imaging raw raster light from the PMTs. The two signals are compared to generate an error signal which is used to increment or decrement the PMT gain.

This known technique suffers from a number of disadvantages. First, the cathode ray tube (CRT) scanning requires modification to direct light to the PMTs during blanking; second, the telecine gate mechanism used for each film standard requires perforation apertures to be fitted in the gate; third, as a consequence of the film gate perforation apertures there is a risk of veiling flare in the active picture; fourthly any noise in the electrical reference signal will inevitably result in noise in the system and possibly also in the image signal; and fifthly, the gain of the blue signal is particularly prone to error as, in any colour system, the inherent lower level of blue light from the CRT means that the blue CRT gain is highest and thus any errors are accentuated in the blue channel.

IBM Technical Disclosure Volume 5 No. 8 of January 1963 at Pages 110 and III entitled "Calibration of Photographic Scanning Systems" by K.H. Trampel and H.H Jensen discloses a system in which two PMTs are used. first PMT, which views the CRT directly, is used as a reference channel and the second is used to view light transmitted through the film. A derived error signal is used to adjust the gain of the second PMT. However, the system assumes that only the second PMT drifts and so any correction made is prone to error as the first PMT drifts. In addition to this disadvantage, the IBM system suffers from all the disadvantages of the prior Cintel system discussed above as well as requiring an additional PMT in each colour channel.

The invention aims to provide an improved method and apparatus for drift compensation and so ameliorate or eliminate the effects of drift. In broad terms, the invention resides in use of a virtual reference value to compensate for drifts in gain of a photoelectric cell.

In one aspect of the invention the reference is not an electrical signal but a target value set by the system software. This has the advantage that the reference is absolutely stable and not subject to the problems of noise identified in the prior art. Preferably the target value is one of a plurality of target values determined according to the transfer characteristic of the cell.

More specifically there is provided a method of correcting for drift in a photoelectric cell, the cell converting incident electromagnetic radiation from a first source into an electrical signal and the magnitude of the signal being determined by a predetermined gain; comprising: deriving a reference value; illuminating the photoelectric cell with radiation from a source to produce an output signal; comparing the output signal with the reference value to produce a comparison result; and adjusting the gain of the photoelectric cell if the comparison result is not within a predetermined allowable range; characterised in that the reference value is a virtual target value generated by a controller controlling the photoelectric cell.

This aspect of the invention also provides apparatus for correcting for drift in a photoelectric cell, the cell converting incident electromagnetic (e/m) radiation from a first source into an electrical signal and the cell having a variable gain comprising: means for deriving a reference signal; an e/m source for illuminating the photoelectric cell; a comparing means for comparing the output of the cell with the reference signal to produce a comparison result; and means for adjusting the gain of the cell if the comparison result is not within a predetermined range; characterised in that the reference signal deriving means comprises means for generating a virtual target value in a controller controlling the photoelectric cell.

Another aspect of the invention resides in the use of a stable independent light source arranged adjacent the photoelectric cell which can illuminate the cell window to produce an output which can be compared with a reference. The PMT gain is varied, if necessary, based on the results of the comparison.

More specifically there is provided apparatus for correcting for drift in a photoelectric cell, the cell converting incident light from a first light source into an electrical signal, and the cell having a variable gain, comprising: means for deriving a reference signal; a stable light source for illuminating the photoelectric cell, the stable light source being independent of the first light source and adjacent the photoelectric cell; a comparing means for comparing the output of the cell with the reference signal to produce a comparison result; and means for adjusting the gain of the cell if the comparison result is not within a predetermined allowable range.

Apparatus embodying the invention has the advantage that the independent stable light source avoids modification of the cathode ray tube scanning when used in a flying spot scanner, and avoids the problems with high PMT blue gain present in the prior art. By arranging the independent light source adjacent the photoelectric cell, the need for perforation apertures in a flying spot scanner is avoided.

The invention also provides a flying spot scanner for converting image stored on film into video signals, comprising a cathode ray tube for providing a scanning raster, a film gate through which film to be scanned is passed, a photoelectric cell for receiving light modulated by the film, and apparatus for correcting for drift in the photoelectric cell as defined above.

The invention further provides a method of correcting for drift in a photoelectric cell, the cell converting incident light from a first light source into an electrical signal and the magnitude of the electrical signal being determined by a predetermined gain; comprising: deriving a reference signal; illuminating the photoelectric cell with light from a stable light source to produce an output signal, wherein the independent stable light source is arranged adjacent the photoelectric cell; comparing the output signal with the reference signal to produce a comparison results; and adjusting the gain of the photoelectric cell if the comparison result is not within a predetermined allowable range.

The invention also provides a method of correcting for photoelectric cell drift in a flying spot scanner having at least one photoelectric cell arranged to receive light from a cathode ray tube modulated by a film image being scanned, the method comprising applying the method set out above, wherein the light from the independent stable light source is blanked during scanning of active picture information on the film by the CRT scanning raster.

Preferably the independent stable light source is a light emitting diode (LED) this has the further advantage that it has substantially no thermal lag, thus avoiding flare in the active picture.

Preferably a separate LED is provided for each of the Red, Green and Blue photoelectric cells in a film scanner. The LEDs are operated during the frame blanking interval to avoid interference with the modulated light from the film incident on the cell window.

An embodiment of the invention will now be described, by way of example only, and with reference to the accompanying drawings, in which:

Figure 1 is a simplified block diagram of a known flying spot scanner previously described; Figure 2 is a graph showing the transfer characteristic of a photomultiplier tube; Figure 3 is a graph illustrating PMT drift with time; Figure 4 is a schematic block diagram of an embodiment of the invention; Figure 5 is a partial cross-section of a photomultiplier tube and LED embodying the invention; Figure 6 is a timing diagram showing when the LED is illuminated; and Figure 7 is a flow diagram showing the main steps executed by software routines controlling the gain of the PMT of Figure 4.

We have appreciated that the light output from a Cathode Ray Tube (CRT), for a given beam current does not change very significantly over a few tens or hundreds of hours. Its stability is very good unless there is a problem with the stabilising of drive voltages controlling it. In practice, with modern technology, these problems do not arise. Thus, the need to compensate for CRT light output variation is significantly lower than the need to compensate for photomultiplier tube (PMT) gain variations. It follows from this that it is not necessary to use CRT light as the source for comparison with a reference to try and measure the absolute level of the signal being derived to determine the system gain drift correction needed. By using a light source other than the CRT, two of the problems identified in the prior art are eliminated: modification of CRT scanning and high error levels in the blue channel due to low blue light from the CRT.

Thus, an independent stable light source may be used to excite the PMT to monitor its system value on a periodic basis. Gain compensation can then be made to the PMT to correct any undesired changes. We have also appreciated that such an independent stable light source is ideally placed adjacent the PMT window on which light is incident to bypass the entire light path from the CRT to the PMT. In other words, by placing the light source on the PMT side of the film gate, the problem of modifying the film gate to include perforation apertures no longer arises.

We have further appreciated that a stable light source with a fast response time, that is having no thermal lag, will overcome the problem of flare in the active picture if the light source is only illuminated during blanking intervals, that is, not when the active picture is being scanned.

A light emitting diode (LED) is a light source which meets all these requirements.

The only disadvantage of the prior art system of GBA-2,215,551 not addressed by the use of an LED proximate the PMT window is that of reference signal noise. We have appreciated that noise is inherent in any electrical reference signal, and there will also inevitably be other uncertainties such as reference signal drift or variation due to machine operation, caused for example by magnetic fields. We have appreciated that this problem may be overcome by generating an arbitrary virtual reference value which can be generated as a target value by the software controlling operation of the scanner. Thus, instead of comparing the measured value to a physical reference, the measured value is compared to a virtual reference, giving an absolute certainty about the comparison; any difference must be due to a movement in the measured value.

We have also appreciated that an implementation of drift gain control in accordance with any inventive aspect discussed above should advantageously allow operator variation of the gain, in addition to drift gain control. This is particularly the case in a telecine embodiment in which the variation of gain of the PMT is a necessary attribute as discussed above. By putting more gain in the PMT, less gain can be used in the electronic processing of a telecine, or vice versa, thereby helping to maintain a high signal to noise ratio at the telecine output dependent upon the type of film being used. The embodiment discussed below allows such gain variation in addition to the gain drift control.

The gain of a PMT system having a head amplifier at the output of the PMT could be adjusted by varying the gain of the head amplifier. However, we have appreciated that, in a telecine embodiment, this would not be appropriate. A telecine head amplifier is usually a transimpedance amplifier which operates with inputs of the order microamps and outputs in the range of millivolts. Any controls signals, such as gain control signals, brought near such a low signal amplifier stage would introduce electrical noise and drift. Accordingly, in telecine embodiment the gain of the PMT itself is adjusted to control drift, rather than the head amplifier.

Turning now to Figure 4, the system shown illustrates the use in the application of an LED light source and virtual reference signal in a single colour channel or film scanner. It will be appreciated that each of the Red, Green and Blue colour channels are treated similarly and only one is described for convenience.

Thus, the system includes a CRT 10 light which scans film 120 as it passes through a film gate 20. Modulated light from the film passes to PMT 30. It is to be understood that the light path between the CRT and the PMT will include imaging optics and dichroic mirrors. These are standard and omitted for simplicity. The output from the PMT is amplified by head amplifier 40 and then converted to a digital signal by analog-to- digital converter 50 under the control of system clocks and timing pulses. The output from the PMT provides the video output 55 for further processing, where the input to the PMT is modulated light from the film.

As explained previously, an LED 110 is mounted adjacent the PMT. Preferably, this is mounted in the cell box (not shown) which contains the dichroic filters which split light from the film into separate colour components.

The LED is mounted such that it can illuminate directly the PMT window through which incident light is collected.

In the embodiment, a red LED is used for the red PMT as the position of the LED is such that the light from the LED is injected prior to the Red trim filter. A green LED is used for the green PMT for the same reason but a green LED is also used for the blue PMT as there is no blue trim filter. In the general case any LED will suffice if the PMT is sensitive to the visible light spectrum, otherwise the LED is chosen to conform to the frequency sensitivity of the PMT it illuminates.

In an alternative embodiment, instead of using a red LED injecting its light into the red PMT via a red filter, the red filter is repositioned and a green LED is used directly adjacent the red PMT window. Similarly, the green LED injects light directly on to the green PMT window (i.e. not via a green filter). The blue PMT also has the green LED directly injecting on to the blue PMT window. A green LED is preferred as it has a high light output stability, and does not suffer from extraneous wideband output such as infrared which would be detected by the PMT.

The output from the PMT, as well as forming the video signal during active scanning is passed to a comparator 60. The comparator 60 receives the PMT output both when there is active scanning and when the LED 110 illuminates the PMT. The comparator compares the two signals under the control of a computer 70 and outputs the result of that comparison to an interface 80. The interface 80 is also controlled by the computer which communicates with the remainder of the film scanner and controls the timing of all operations. In essence this is well known. For example the computer controls line drive pulses, frame drive and frame blanking pulses.

The line drive and frame drive are both passed to the interface and are used to ensure that a control signal to a PMT drive 90, which will vary the gain of the PMT does not occur during active scanning and that changes are made progressively over a number of frames to avoid noticeable steps in the video level. The provision of the line drive and frame blanking controls to the LED driver 100 enable control of the LED to illuminate the PMT window only during the frame blanking intervals. During other, non active, picture periods such as line blanking and vertical flyback periods it is desirable to blank the LED so that black level line clamps are not disturbed.

The comparator 60, described further later, ensures that the gain stabilisation is achieved with the PMT over a range of PMT gain values that may be selected by the operator.

Turning now to Figure 5, the manner in which the LED is mounted is illustrated in greater detail. The tube of the photomultiplier is mounted in a cell box 122.

The cell box includes an aperture 124 or trim filter behind which is arranged the PMT window 126 through which incident light is received. The LED 110 is mounted on an arm 128 bolted to the cell box; the arm being angled towards the PMT such that light from the LED can fall on the PMT window 126.

Figure 6 shows therelative timing of the LEDs. The frame blanking interval is illustrated at 130 and it can be seen that the LED is pulsed a short time after commencement of frame blanking when the frame blanking level goes low. A series of three LED pulses 132 are initiated a short time after the onset of frame blanking, typically 150pS although the timing may be varied. The number of LED pulses may also be varied. Preferably each pulse lasts 20iS and pulsing occurs on three consecutive video lines with one pulse on each line. In this manner the LED 110 is controlled using field and line pulses.

Turning now to Figure 7, the manner in which the system calculates the reference voltages and applies the comparison will be described with reference to the flow chart.

In telecine machines such as the URSA flying spot telecine, an automatic shading operation is preferred. This is described in our earlier application W089/01539. In the auto shading process a map of the CRT face plate is produced and a correction factor derived for each pixel according to its response in an open gate position. This correction map is, of course, created using light incident on the PMT. This process is also referred to as alignment. The system utilises this alignment procedure to establish a virtual reference set as a value created in the software of the PMT drive alignment. The virtual reference is actually a target value for the system gain for a different input stimulus condition.

In practice, a film scanner is required to work with a number of different film types such as negative, inter positive and print films. Each has a different film density and the scanner is required to achieve maximum signal-to-noise performance with each. Thus, to operate on the basis of a virtual reference, the target value must be adjustable. As has been mentioned with reference to Figure 2, the current gain: anode voltage transfer characteristic of the PMT is substantially logarithmic over its operating range. From this it can be seen that by exercising and measuring four different stimulus points for the system, corresponding to a range of density values, the transfer characteristic of the PMT can be approximated by interpolating linearly between the operating points. This gives a reasonably good characterisation of the PMT. The number of stimulus points may be varied. In fact, it is preferred to use seven reference points between the maximum and minimum PMT drive voltages (gain) during the alignment process to produce a look up table. The use of more reference points produces an enhanced performance. If too few points are used, the stabilisation system may visibly pull to a different gain value when the operator selects a gain because the look up table would be insufficiently defined.

For simplicity, the following description will refer to four reference values only.

In Figure 7, the four reference values are known from the alignment procedure on start up and intermediate values derived by interpolation. The system software at 200 retrieves the current control value of the PMT from the system control desk and derives the appropriate reference value from the look up table 210. The value is output to an adder function 220. It will be appreciated that the reference value is simply a software value and, as such, immune to noise problems associated with a conventional reference signal.

The PMT control value 200 is input also to an adder function 230 and forms a drive voltage input to the PMTs at 240. The output from the PMT with the LED 110 switched on is input, once converted to a digital value, to a further adder function 250 which has as its other input, the inverted output from the adder 220. This value, on start up will simply be the calculated reference value. The output of the adder 250 is the difference between the actual PMT output and the reference value. This is examined at 260 and 270 to see if it is a positive or negative result. If it is negative then the reference offset is increased at 280 and if it is positive the reference offset is decreased at 290. The reference offset is then examined at 300 and 310 to determine whether it is within maximum positive and negative thresholds. If it is, then the offset is passed (at 320) to a positive input of adder function 220. Thus, for small variations between reference voltage and measured voltage the reference is adjusted. However, if the reference is outside the threshold values, the PMT automatic gain control is offset by decreasing if the reference offset is above the threshold, at step 330, or increasing it at 340 if the reference offset is below the negative threshold. The Automatic Gain Control (AGC) offset is added to the gain control value from the control desk at adder function 230 and applied to the PMT as a fresh gain value.

The process described takes place only when the PMT LED is illuminated.

The initial calibration of the reference values for the look up table is determined by setting a target value into the comparator on start up when alignment first takes place. It will be appreciated that the comparator performs functions 250, 260 and 270 in Figure 7. During start up, no offsets are applied at 220 or 230 in Figure 7. A drive voltage is applied to the PMT and an arbitrary reference, a virtual reference, is applied to the comparator (functions 250, 260 and 270 in Figure 7). The comparator tells whether the first guess at a reference value was high or low enabling adjustment to be made in the manner described. The correct value is therefore achieved by the iterative process. The process is first performed when the CRT is aligned at maximum PMT drive voltage to set a first value and the second value set during alignment at minimum PMT drive voltage. Two other values between the maximum and minimum are then determined in the same manner to produce the four known points shown at 210 in Figure 7.

The routine of Figure 7 is disabled when the EHT supply to the CRT is off. This avoids the possibility that an enormous AGC offset is developed.

In summary, the embodiment described overcomes the disadvantages of the prior art by providing an independent stable light source adjacent the PMT by comparing PMT output during illumination by that light source with a virtual reference generated by the system software. It has been found that a system embodying the invention has reduced drift from about 18% to 1%.

The invention has been described with reference to a PMT in a film scanner. However it should be understood that it is applicable to any type of photoelectric cell in any application where drift is a problem. These applications include photometry, Pollution detection, scintillation counters, astronomy and x-ray detection.

Claims (38)

Claims

1. Apparatus for correcting for drift in a photoelectric cell, the cell converting incident electromagnetic (e/m) radiation from a first source into an electrical signal and the cell having a variable gain comprising: means for deriving a reference signal; an e/m source for illuminating the photoelectric cell; a comparing means for comparing the output of the cell with the reference signal to produce a comparison result; and means for adjusting the gain of the cell if the comparison result is not within a predetermined range; characterised in that the reference signal deriving means comprises means for generating a virtual target value in controller controlling the photoelectric cell.

2. Apparatus according to claim 1, wherein the photoelectric cell is a photomultiplier tube.

3. Apparatus according to claim 1 or 2, wherein the gain of the photoelectric cell is of variable levels, and the drift correction is operable at any said level.

4. Apparatus according to any of claims 1 to 3, wherein the reference value is a target reference value for a given input condition.

5. Apparatus according to claim 4, wherein the target reference value is selected from an interpolation of the transfer characteristic of the cell.

6. Apparatus according to claim 5, wherein the interpolation is used to derive a lookup table of reference values.

7. A flying spot scanner for converting image stored on film into video signals, comprising a cathode ray tube for providing a scanning raster, a film gate through which film to be scanned is passed, a photoelectric cell for receiving light modulated by the film, and apparatus for correcting for drift in the photoelectric cell according to any preceding claim.

8. A method of correcting for drift in a photoelectric cell, the cell converting incident electromagnetic radiation from a first source into an electrical signal and the magnitude of the signal being determined by a predetermined gain; comprising: deriving a reference value; illuminating the photoelectric cell with radiation from a source to produce an output signal; comparing the output signal with the reference value to produce a comparison result; and adjusting the gain of the photoelectric cell if the comparison result is not within a predetermined allowable range; characterised in that the reference value is a virtual target value generated by a controller controlling the photoelectric cell.

9. A method according to claim 8, wherein the gain of the photoelectric cell is of variable levels, and the drift correction is operable at any said level.

10. A method according to claim 8 or 9, wherein the reference value is a target reference value for a given input condition.

11. A method according to claim 10, wherein the target reference value is selected from an interpolation of the transfer characteristic of the cell.

- 19

12. A method according to claim 11, wherein the interpolation is used to derive a lookup table of reference values.

13. A method according to any of claims 8 to 12, wherein the step of comparing comprises: varying the reference signal offset if the output signal and the reference are not equal; comparing the varied signal offset to an allowable threshold; and adding the varied offset to the reference signal if the offset is within the threshold.

14. A method according to claim 13, wherein if the reference offset is outside the threshold the reference signal is not added to the reference signal, and the cell gain offset is varied.

15. Apparatus for correcting for drift in a photoelectric cell, the cell converting incident electromagnetic (e/m) radiation from a first e/m source into an electrical signal, and the cell having a variable gain, comprising: means for deriving a reference signal; a stable e/m source for illuminating the photoelectric cell, the stable e/m source being independent of the first e/m source and adjacent the photoelectric cell; a comparing means for comparing the output of the cell with the reference signal to produce a comparison result; and means for adjusting the gain of the cell if the comparison result is not within a predetermined allowable range.

16. Apparatus according to claim 15, wherein the independent stable e/m source is positioned such that it can illuminate the photoelectric cell along a light path - distinct from that by which the first light source illuminates the cell.

17. Apparatus according to claim 15, wherein the photoelectric cell is a photomultiplier tube.

18. Apparatus according to claim 15, 16 or 17, wherein the independent stable e/m source has substantially no thermal lag.

19. Apparatus according to claim 15, 16, 17 or 18, wherein the independent e/m source is a light emitting diode.

20. Apparatus according to any of claims 15 to 19, wherein the reference signal is a target reference value for a given input condition.

21. Apparatus according to claim 20, wherein the target reference value is selected from an interpolation of the transfer characteristic of the cell.

22. Apparatus according to any of claims 15 to 21, wherein the independent stable e/m source is mounted on a cell box housing the photoelectric cell and angled to illuminate the cell window.

23. A flying spot scanner for converting image stored on film into video signals, comprising a cathode ray tube for providing a scanning raster, a film gate through which film to be scanned is passed, a photoelectric cell for receiving light modulated by the film, and apparatus for correcting for drift in the photoelectric cell according to any of claims 15 to 22.

24. A flying spot scanner according to claim 23, comprising a controller for controlling the independent stable e/m source to illuminate the photoelectric cell during the blanking interval between active film frames.

25. A flying spot scanner according to claim 24, wherein the e/m source is switched after a delay of about 150gS after onset of the frame blanking interval.

26. A flying spot scanner according to claim 24 or 25, wherein the e/m source is switched a plurality of times during the blanking interval.

27. A flying spot scanner according to claim 26, wherein each switching of the e/m source occurs on a fresh video line.

28. A method of correcting for drift in a photoelectric cell, the cell converting incident electromagnetic (e/m) radiation from a first e/m source into an electrical signal and the magnitude of the electrical signal being determined by a predetermined gain; comprising: deriving a reference signal; illuminating the photoelectrical cell with e/m radiation from a stable light source to produce an output signal, wherein the independent stable e/m source is arranged adjacent the photoelectric cell; comparing the output signal with the reference signal to produce a comparison result; and adjusting the gain of the photoelectric cell if the comparison result,is not within a predetermined allowable range.

29. A method according to claim 28, wherein the step of deriving a reference signal comprises deriving a target value for a given input condition.

30. A method according to claim 28 or 29, wherein the step of comparing comprises: varying the reference signal offset if the output signal and the reference are not equal; comparing the varied signal offset to an allowable threshold; and adding the varied offset to the reference signal if the offset is within the threshold.

31. A method according to claim 30, wherein if the reference offset is outside the threshold the reference signal is not added to the reference signal, and the cell gain offset is varied.

32. The method of correcting for photoelectric cell drift in a flying spot scanner having at least one photoelectric cell arranged to receive light from a cathode ray tube modulated by a film image being scanned, the method comprising applying the method of any claims 28 to 31 wherein the light from the independent stable light source is blanked during scanning of active picture information on the film by the CRT scanning raster.

33. A method according to claim 32, wherein the independent stable light source is blanked during line blanking intervals.

34. A method according to claim 32 or 33, wherein the CRT raster is blanked when light from the independent stable light source is incident the photoelectric cell.

35. Apparatus for correcting for drift in a photoelectric cell. substantially as herein described with reference to Figures 4 to 7 of the accompanying drawings.

36. A flying spot scanner substantially as herein described with reference to Figures 4 to 7 of the accompanying drawings.

37. A method of correcting for drift in a photoelectric cell substantially as herein described with reference to Figures 4 to 7 of the accompanying drawings.

38. A method of correcting for photoelectric cel ' 1 drift in a flying spot scanner having at least one photoelectric cell, substantially as herein described with reference to Figures 4 to 7 of the accompanying drawings.