GB2059705A - Electronic noise suppression - Google Patents

Abstract

Charge-coupled devices (c.c.d.) delay lines, when used as temporary stores of information in electrical form, exhibit a tendency to distort the stored information. Each signal is affected similarly so the distortion is called fixed pattern noise. The invention overcomes this problem by a circuit arrangement which apparently eliminates the distortion. Alternate signals of a succession of signals are inverted by a variable-sign amplifier (2) before storage in a c.c.d. delay line (6) and re-inverted in a second variable- sign amplifier (9) after removal from the store (6); the result is to reverse the distortion on sequential signals and thus significantly reduce the effect of the distortion on any consecutive pair of signals. The main use of the invention is in infra- red imaging apparatus, providing a visible display of a scene viewed by infra-red detectors. (Fig. 1). <IMAGE>

Description

SPECIFICATION Electronic noise suppression This invention relates to the provision of means for the suppression of electronic noise, and is concerned in particular with means for suppressing the fixed pattern noise generated in charge-coupled delay line devices.

Noise in a signal transmission system may be defined as that part of the system's output which is "not-signal", or as that part which interferes with the signal. "Fixed pattern noise" is the term used to describe the noise superimposed on a signal temporarily stored in a charge-coupled device delay line (a charge-coupled device a c.c.d.-is essentially an assembly of semi-conductor cell devices for the temporary storage of electrical signals) as a result of the undesirable but unavoidable leakage of the stored charge out of or into the c.c.d.'s individual cells. Not only can this leakage result in the overall level of charge being altered, but in addition the individual cells, having differing leakage characteristics, can cause non-uniform distortion of the charge (signal) stored therein.If this distortion could be minimised, the use of c.c.d.'s could be made more attractive, and even extended to cover many applications for which they are presently. considered to be unsuitable.

One field of activity where c.c.d.'s are used is in the formation of delay lines such as those employed in infra-red imaging apparatus, where an array of infra-red (IR) detectors viewing a scene produces electrical signals which are processed into a visible scanned (TV-type) display of the scene. One problem of interfacing a scanned display system with an IR detector array is that arising from the need to provide a scan conversion system. Whereas the IR detector array views the scene more or less instantaneously, the display picture is built up, by a scanning process, over a relatively much longer period of time; the difficulty lies in the conversion of the array's output into a form more suitable for the display's input-and more specifically in the need to store the IR detector array's output until it is convenient for the display system to accept it.

The use of c.c.d's as delay lines (or temporary stores) for large numbers of discrete pieces of electrical information (in this case the output of the IR detectors) is a convenient way of overcoming this problem-but it would be more so if it were not for the problems of the c.c.d.-generated fixed pattern noise, which seriously affects the quality of the image displayed on the screen.

One use of this invention to provide means for, and a method of, overcoming the problem of fixed pattern noise generated by the c.c.d.

delay line by using a circuit arrangement which apparently eliminates the noise from the required signals.

Accordingly, in one aspect this invention provides apparatus for the temporary storage of successive electrical signals in a store which distorts each signal in a similar way, wherein there is provided: a) means for inverting the signal before it is entered into the store; b) means for re-inverting that signal after it is removed from the store; and c) means for determining that one signal shall be so inverted and subsequently reinverted, and that the sequential signal shall not be so inverted and subsequently re-inverted; With the result that in operation any distortion superimposed on the signal while in the store is reversed on sequential signals so that the effect of any such distortion superimposed on any consecutive pair of signals is significantly reduced.

The expression "signal" is used herein to include a set of signals.

The store is preferably a delay line of the charge-coupled device type.

The means for inverting the incoming signal is preferably a variable-sign amplifier as hereinafter defined.

The means for re-inverting the signal output from the store is preferably also a variablesign amplifier.

Amplifiers of electronic signals normally provide a replica of the input which, ideally, is increased in size (amplitude) only but is, in all other respect, identical with the input. However, by using suitable input/output circuitry it is possible to make an amplifier the output of which, although otherwise identical with the input (except in amplitude), is actually inverted with respect to the input. Again, it is possible to make amplifiers the operation of which can be changed to produce either a normal (non-inverted) amplified output or an inverted amplified output, the change being effected by change-over means operated by external control signals. This type of amplifier is known as a "variable-sign amplifier".A variable-sign amplifier is basically an amplifying device which has two alternative input/ output circuits each of which may be selected by switching means so as to provide either an output which is inverted with respect to the input or an output which is not so inverted.

In a somewhat more limited manner, it will be seen that the apparatus of the invention may be defined as a charge-coupled delay line device constituting electronic noise suppression apparatus and comprising:~ a) a first variable-sign amplifier the output of which is used as the input to b) a charge-coupled device (c.c.d.) in which the output of the first variable-sign amplifier is stored until extracted and used as the input for c) a second variable-s#ign arnplifier having the same sign as the first variable-sign amplifier, and d) means for changing the sign of the amplifier pair such that, in operation, over a succession of sign changes (that is, at least two), the arithmetic means (average) of any fixed pattern noise generated by charge leak age in the c.c.d. is significantly reduced.

Each variable-sign amplifier is advantage ously an operational amplifier associated with a set of logic-controlled analogue switches connecting either inverting or non-inverting circuit configurations to the amplifier, and wherein the factor controlling the position of the switches is the presence (or absence) of a signal derived from a regularly-occuring elec trical activity: Each operational amplifier (of the above general type) is preferably of the type com monly known as a wideband amplifier-the type which will accept sign#als with a range of frequency of from to 5 MHz (commonly known as a video amplifier).

Each variable-sign amplifier is conveniently of the above general type in which an opera tional amplifier is #arranged to provide an amplification factor (gain) equal to 1 (i.e. the output is equal to the input).

As stated above, the second variable sign amplifier is conveniently of the same sort as the first, but in some circumstances it may be advantageous for the first amplifier to have#an amplification factor (gain) equal to 1, and for the second amplifier to have an amplification factor (gain) greater than 1 in order to provide additional signal strength for the feedback configuration upon which the operation sign -of the -amplifier depends.

In particular, a variable-sign wideband am plifier of this type has been constructed using operational amplifier circuits of the type known as HA 2525 manufactured by Harris Semiconductor. In addition, operational amplifiers from the LF 155, 156 series of circuits manufactured by National Semiconductors have# also been found to be suitable for this purpose.

The means for changing the sign of the amplification produced by the two amplifiers is preferably a set of logic-controlled analogue gates or switches connecting either the invert ing or the non-inverting circuit configuration to the amplifier. The factor controlling the position of the switches (open or closed) is the presence (or absence) of a signal derived from some regularly-occuring electrical activiW-for example, the alternation of the mains supply voltage.

Where the charge-coupled device for sup pressing electronic noise is used in a system providing a visible image of an infra-red scene on a scanned display, the factor initiating switching from one circuit configuration to anther may be a signal derived from the field-synchronising pulses of the scanned display.

A full picture (or frame) of a scanned display normally consists of two sets (or fields) of inter-laced lines of picture information, each field occurring at regular intervals (which interval is usually the frequency of the mains supply); the field of lines carrying the information which comprises one-half (in alternate line strips) of the frame must be synchronised with the other field comprising the other half of the frame to appear at the correct time, and the signal arranged to provide this synchronisation is a pulse derived from the mains frequency, referred to as the field-synchronis ing pulse.

Where the invention is used in an infra-red detection system providing a visible light scanned display, another convenient method of getting the signal required to operate the -analogue gates or switches is to derive it from the synchronising pulses of the IR scanning system (which pulses ensure that its electrical outputs are all transmitted at the same time).

Further methods of obtaining this signal might include a signal produced within the c.c.d. itself, or a suitable 'clock-frequency' signal from some other outside source; but both of these sources have the disadvantage that they would not be synchronised to either the scanning of the display or the IR scanning device, and thus would effectively 'free-run', with associated problems of synchronisation.

In another aspect this invention provides a method of suppressing noise generated in electronic apparatus for the temporary storage of successive signals, the storage means distorting each signal in a similar way, in which method a first signal is inverted before it is stored in, and re-inverted after it is removed from, the storage means, with the result that any distortion superimposed thereon is reversed, while the sequential signal is stored without inversion and re-inversion, so that the effect of any distortion superimposed on any such pair of signals is significantly reduced.

The means used to implement the above mentioned method of electronic noise sup pression will be obvious from the foregoing description of the apparatus of the invention, and it is not considered that any further description of the modus operandi is necessary.

As mentioned above, one application of this invention is in a scan conversion system of an infra-red imaging apparatus displaying a visi ble image of an infra-red scene by means of a scanned display, which system stores all the infra-red detector outputs until such time as they are required for display.

A convenient method of doing this uses a c.c.d delay line as a store. The problems caused by the inherent leakage characteristic of each storage cell in the c.c.d. may be overcome by using the apparatus and method of the invention, as is now described.

A variable-sign amplifier is included both in the input and in the output paths of the store, and it is arranged that the sign of the amplification be changed at convenient intervals, the most convenient interval being every new field of the display. This ensures that on alternate fields of the picture information the signal is stored in an inverted form. The effect of this is to add or subtract the fixed pattern noise derived from the delay line to or from the required signal so that the storage device's output signal to the display is either "(signal + noise)" or "(signal - noise)" displayed alternately.

Provided that the display's "frame rate" (that is, the number of frames displayed per second) is sufficiently high so as to be faster than the "flicker rate" of the eye (that is, the highest rate at which the human eye can perceive fluctuations of brightness), the fixed pattern noise (being reversed on every display field) will be effectively eliminated by the integration performed by the persistence of vision of the observer's eye. There may be some residual fixed pattern noise occuring at half the frame rate, because the sign reversal signal is derived from the field synchronising pulses, but this will be at a significantly lower level than the patterning in the original signal.

The effects of the residual fixed pattern noise may be further decreased by the afterglow of the phosphors used in the screen, if the luminescence decay time is comparable to that of the frame time.

The invention will now be described, though only by way of illustration, with reference to the accompanying drawings, in which: Figure 1 shows in diagrammatic circuit form one embodiment of an apparatus of the invention; Figure 2 shows in diagrammatic form the time relationship between the input to the apparatus shown in Fig. 1 and its corresponding output; Figures 3(a) to 3(g) show in graphical form various electrical activities taking place in one embodiment of an apparatus of the invention from time to time.

Figure 4 shows in diagrammatic circuit form one embodiment of the variable-sign amplifier used in the invention.

Like references refer to similar features on the drawings.

Referring to Fig. 1, an input terminal 1 is connected to a first variable-sign amplifier 2 having a "write" sign control signal input terminal 3, the amplifier's output terminal 4 being connected to the input terminal 5 of a c.c.d. delay line (store) 6. The output terminal 7 of the store 6 is connected to the input terminal 8 of a second variable-sign amplifier 9 having a "read" sign control signal input terminal 10, this amplifier's output being connected to an output terminal 11.

In operation, an analogue input signal (in this particular embodiment, an IR detector output signal) is applied to the input terminal 1 of the first variable-sign amplifier 2, and a write sign control signal is then applied to terminal 3. This "write" signal causes the amplifier 2 to pass its output at terminal 4 to the store 6, via input terminal 5.

It also causes alternate output signals from amplifier 2 to be inverted with respect to their corresponding input when so passed.

After the required delay has been imposed by storing the signal in the store 6, the output signal, which is now the required signal plus the fixed pattern noise acquired in the store 6, is passed from output terminal 7 to the input terminal 8 of the second variable-sign amplifier 9. This amplifier's output is applied to an output terminal 11 under the control of a "read" sign control signal applied to terminal 10. This "read" signal causes the amplifier 9 to pass its output to the output terminal 11, and also causes alternate amplifier output signals to be inverted with respect to the corresponding input signals when so passed.

The circuit diagram of the variable-sign amplifiers is shown in Fig. 4, and its operation described below.

Fig. 2 shows in diagrammatic form the time relationship between an input signal to, and the corresponding output signal from, the store 6 and its associated amplifiers 2 and 9.

The upper part of Fig. 2 shows a representation of the output signals (lines 1-6, etc.) from the detector array of an IR viewing apparatus of the type in which a number of IR detector elements have the external thermal scene presented thereto by a rotating multi faceted mirror assembly.

The presentation of the external thermal scene may be carried out in a variety of ways, but in a simple embodiment, the axis of rotation of the mirror assembly "nods" through a sufficient elevation angle to scan the elevation angle of the required view.

In another version, the facets of the mirror are arranged at different angles of elevation (or depressionj so as to achieve the same result during one complete rotational sweep.

Similar devices are embodied in such diverse equipment as airborne FLIR systems, thermal imaging surveillance systems and weapon aiming systems such as anti-tank or antiaircraft gun sights.

The lower part of Fig. 2 shows that after passing these signals (the detector array outputs) to the store 6 there is provided a sequentially-consecutive output with the required delay for the display.

The graphs of Figs. 3(a) to 3(g) shows various electrical activities associated with the store 6 and its amplifiers 2 and 9. Fig. 3(a) represents the fixed pattern noise in a particular c.c.d. store 6, the various "steps" up and down being an exaggerated representation of the leakage occuring in individual cells. Fig.

3(b) depicts a first analogue signal derived by an IR detector from a thermal scene (corresponding to one particular line of a first com plete picture, or frame), the varying amplitude of the signal being caused by the varying electrical output of the IR detector according to variations of temperature of the thermal scene. Fig. 3(c) shows the same signal after having passed through the first variable-sign amplifier 2 (operating, here, in the positive, or normal sense) and being then "written" into the c.c.d. store 6, with the various "steps" being the store's approximation to the original input signal (as shown by the thin dashed line) before the fixed pattern noise has taken effect.

Fig. 3(d)(i) represents the effect of the fixed pattern noise on the (temporarily) stored signal when it is held in the particular store 6 with the fixed pattern noise shown in Fig.

3(a); as can be seen by comparing it with the superimposed thin dashed line of the original signal and with Fig. 3(c), it departs substantially from the originally stored charge pattern of Fig. 3(c), because the fixed pattern noise of the store 6 has now been added to the required signal-that is, it is "(signal + fixed noise)".

Fig. 3(d)(ii) shows the actual output signal (including the fixed pattern noise) "read" out to the display from the apparatus of the invention; it may readily be seen that it is now considerably distorted as compared to the original input signal (shown in dashed line).

Fig. 3(e) shows "written" into the store 6 a second IR detector analogue signal similar to that shown in Fig. 3(c) and corresponding to the same line of a second frame, but the first variable-sign amplifier 2 is now operating in the negative sense, inverting its output signal with respect to the original input signal. It can be seen from Fig. 3(f)(i) that the effect of the fixed pattern noise on the (temporarily) stored signal when it is held in the store 6 is to cause it to depart substantially from the originally stored charge pattern of Fig. 3(e), being now "(signal - fixed noise)". The output signal shown in Fig. 3(f)(ii) "read" out to the display only bears a passing resemblance to the original input, as may be seen by comparing it with the original input signal (shown dashed).

In Fig. 3(g), the output signal of Fig. 3(d)(ii) is shown in pecked line superimposed on the output signal of Fig. 3(f)(ii) shown chaindotted, with an additional curve in solid line which in fact shows an "average" of the two signals, this being the effect of viewing the two output signals "together" which is what happens when the image is viewed by a human eye, the persistence of vision being such that it is impossible to see each line of each frame separately, only the overall effect being observable. As may be seen from Fig.

3(g), the "average" or mean curve (shown in solid line) is a fair approximation to the original input signal of Fig. 3(b).

The variable sign amplifiers used in this invention are not available as manufacturer's stock items, but may be made by building two alternative circuits around a standard "operational amplifier" (so-called), one circuit being arranged to provide an inverted output and the other circuit providing a non-inverted output. Each circuit is connected to the operational amplifier by logic-controlled analogue gates or switches operated by the logical presence or absence of external control signals. One embodiment of such an amplifier is now described by way of example only with reference to Fig. 4 of the drawings.

In Fig. 4, an operational amplifier (12) is connected to an input terminal (1) and an output terminal (4) via various circuit components, including resistors of various sorts (RIN), (RF), etc., and four logic-controlled analogue gates or switches Al, A2, A3 and A4. The circuit diagram of the amplifier must be read in conjunction with the adjacent logic "truth table". Note that a logical 1 on the analogue switch control signal input terminals 3(a) or (b) closes the switch, and a logical 0 opens it; also that the logical symbols "A" and "A" are read as "A" and "NOT-A" or "bar-A" respectively (i.e. presence or absence of signal A).

The inverting amplifier circuit configuration is provided as follows:~ Consider signal A at logic 1 (and conversely, A at logic 0) applied to control signal input terminals (3)(a) on switches Al and A3; since A is at logic 1, the switches must be closed, and of low impedance. Signal A (NOT-A) at logic 0 is applied to terminals 3(b) on switches A2 and A4; since A is at logic 0, the switches are open-circuit and of high impedance. The amplifier circuit thus appears in a standard inverting configuration, with gain~ RF= 1, RIN IN if RF = RIN.

The non-inverting amplifier circuit configuration is provided as follows:- Consider signal A at logic 0 (and conversely, A at logic 1): Switches Al and A3 are now open, and switches A2 and A4 are now closed. The amplifier circuit now appears as a standard unity gain or voltagefollower circuit.

Claims (17)

1. Apparatus for the temporary storage of successive electrical signals in a store which distorts each signal in a similar way, wherein there is provided: a) means for inverting the signal before it is entered into the store; b) means for re-inverting that signal after it is removed from the store; and c) means for determining that one signal shall be so inverted and subsequently reinverted, and that the sequential signal shall not be so inverted and subsequently re-inverted; with the result that in operation any distortion superimposed on the signal while in the store is reversed on sequential signals so that the effect of any such distortion superimposed on any consecutive pair of signals is significantly reduced.

2. Apparatus as claimed in claim 1, wherein the store is a delay line of the chargecoupled device type.

3. Apparatus as claimed in either of claim 1 and 2, wherein the means for inverting the incoming signal is a variable-sign amplifier.

4. Apparatus as claimed in any of the preceding claims, wherein the means for reinverting the signal output from the store is a variable-sign amplifier.

5. Apparatus as claimed in any of the preceding claims which is a charge-coupled delay line device constituting electronic noise suppression apparatus and comprising:~ a) a first variable-sign amplifier the output of which is used as the input to b) a charge-coupled device (c.c.d.) in which the output of the first variable-sign amplifier is stored until extracted and used as the input for c) a second variable-sign amplifier having the same sign as the first variable-sign amplifier, and d) means for changing the sign of the amplifier pair such that, in operation, over a succession of sign changes (that is, at least two), the arithmetic mean (average) of any fixed pattern noise generated by charge leakage in the c.c.d. is significantly reduced.

6. Apparatus as claimed in claim 5, wherein each variable-sign amplifier is an operational amplifier associated with a set of logic-controlled analogue switches connecting either inverting or non-inverting circuit configurations to the amplifier, and wherein the factor controlling the position of the switches is the presence (or absence) of a signal derived from a regularly-occurring electrical activity.

7. Apparatus as claimed in claim 6, wherein each operational amplifier is a wideband amplifier.

8. Apparatus as claimed in either of claims 6 and 7, wherein each operational amplifier is an operational amplifier arranged to provide an amplification factor equal to 1.

9. Apparatus as claimed in either of the claim 6 and 7, wherein the first amplifier has an amplification factor equal to 1, and the second amplifier has an amplification factor greater than 1.

10. Apparatus as claimed in any of claims 6 to 9 wherein the charge-coupled device for suppressing electronic noise is used in a system providing a visible image of an infra-red scene on a scanned display, and wherein the factor initiating switching from one circuit configuration to another is a signal derived from the field-synchronising pulses of the scanned display.

11. Apparatus as claimed in claim 6, wherein the factor initiating switching from one circuit configuration to another is a signal derived from the synchronising pulses of an IR scanning system which ensure that all the electrical outputs of such a system are transmitted simultaneously.

12. Apparatus for the temporary storage of successive electrical signals as claimed in any of the preceding claims and substantially as hereinbefore described with reference to the accompanying drawings.

13. A method of suppressing noise generated in electronic apparatus for the temporary storage of successive signals, the storage means distorting each signal in a similar way, in which method a first signal is inverted before it is stored in, and re-inverted after it is removed from, the storage means, with the result that any distortion superimposed thereon is reversed, while the sequential signal is stored without inversion and re-inversion, so that the effect of any distortion superimposed on any such pair of signals is significantly reduced.

14. A method of suppressing noise generated in electronic apparatus as claimed in claim 13 and substantially as hereinbefore described with reference to the accompanying drawings.

15. A scan conversion system of an infrared imaging apparatus displaying a visible image of an infra-red scene by means of a scanned display, in which system the apparatus and/or method as claimed in any of the preceding claims is employed to store all the infra-red detector outputs until such time as they are required for display.

1 6. Apparatus in which successive analogue signals are each temporarily stored in a store, the store being such as to distort each signal in a similar way, the apparatus including means for inverting some but not all the signals before and after storage so that the effect of the distortion on those signals is reversed.

17. Apparatus in which a scene is repetitively scanned and signals derived from each scan are fed to a storage device, characterized by means for inverting the signals derived from alternate scans before and after they are entered in the storage device.