US 20050111661 A1
An apparatus and method for applying anti-copy protection to a raster-scanned video signal are described. A square waveform 52 is added to the vertical synchronisation pulses 50 of the video signal. The effect of the waveform is not apparent until it has been recorded by a video cassette recorder. The anti-copy protected video signal can therefore be viewed normally on a display device at or near to the picture quality intended by the originator of the video signal, but once recorded, exhibits a reduction in picture quality which makes the video signal quite unattractive to view.
1. Apparatus for applying anti-copy-protection to a raster-scanned video signal to produce a modified video signal, the apparatus comprising:
an input for receiving a video signal which is to be modified, the video signal having vertical synchronisation pulses;
modifying means for modifying the vertical synchronisation pulses of a video signal received by the input to produce a modified video signal, the modified signal being viewable on a display device at or near to the quality intended by the originator of the video signal but such that after recording by a video cassette recorder, the average voltage level of a modified vertical synchronisation pulse is shifted towards the blanking level in comparison to the average level before recording, such that the recorded signal exhibits a reduction in vertical stability when played back and viewed on the display device; and
output means for outputting the modified video signal.
The present application relates to a method and an apparatus for modifying a raster-scanned video signal such that the modified video signal will still be viewable on a display device, such as a television screen, but that following recording of the modified video signal, playback of the recorded modified signal will be adversely affected.
The present application provides a way of preventing unauthorised copying of an original video signal, and may be used, in particular, to deter copying onto tape of Pay-Per-View video signals, and video-to-video (tape-to-tape) copying.
Video piracy is a significant problem for broadcasters and distributors of video signals, since the unauthorised copying and distribution of video recordings by pirates can impact drastically on the revenue generated by a broadcaster or distributor through legitimate sales. It is therefore desirable to prevent video pirates from making unauthorised copies of video signals.
This is best achieved by modifying the original video signal such that when it is recorded the recorded signal cannot be satisfactorily played back.
In known protection techniques, the unauthorised recording of the video signal is made less enjoyable to watch by the interaction of the original signal to which the protection has been applied with the electronic components in either the video cassette recorder or the television receiver itself. For example, making an unauthorised copy of the video signal too dark to be viewed satisfactorily may be achieved by adding to the original video signal a pulse which is significantly larger than that part of the signal which carries the picture information. The position at which the pulse is added depends on the way in which the circuits in the television receiver or video recorder process the signal. When the modified signal is processed by the automatic gain control circuits of a video cassette recorder, the amplitude of the signal is perceived as being that of the added pulse and not that of the portion of the signal carrying the useful information. Consequently, the video cassette recorder or television receiver amplifies the received signal by a smaller factor than if the pulse was not present. As a result of this the information-carrying portion of the signal is not therefore amplified enough to be seen satisfactorily when reproduced.
Such methods have however a number of drawbacks. Methods which rely on the automatic gain control of the video cassette recorder, such as adding a large pulse to the signal, tend to result in a modified signal that cannot itself be viewed on a television through the video channel regardless of whether the signal is being or has been recorded.
An alternative protection technique described in International Patent Application WO 01/76240 involves the removal of a small number of horizontal synchronisation pulses from the blanking section of the signal, so that an unauthorised recording of the signal cannot be properly synchronised by the TV receiver on which it is to be played back. As a result, the resulting picture playback can be poor.
A further technique, disclosed in International Patent Application WO 96/31878, relies on inserting a pulse into the colour burst information portion of the signal section of the signal, such that automatic gain control circuits that rely on the average dc level of the colour burst to determine the necessary amplification of the signal, make such amplification too small. An opposing pulse signal having a magnitude sufficient to offset the change in dc level of the colour burst portion caused by the pulse signal, and optionally a second pulse, are inserted somewhere from the last half of the remainder of the back porch of the signal to the end of the start of the picture information portion.
We have found that this technique can be unreliable in practice, and furthermore has the disadvantage that by inserting a pulse into the colour burst part of the signal, the resulting picture quality is detrimentally affected.
Both of these techniques rely on the components employed in the video cassette recorder or the television receiver, and, in some cases, certain video recorders or television receivers may have an arrangement of components that is not susceptible to the adverse picture effects caused by the modified signal. Thus, the modified video signal can still be played without significant detriment to the picture playback and the anti-copy protection applied to the modified signal is rendered useless.
The invention is defined by the independent claims below to which reference should now be made. Advantageous features are set forth in the appendant claims.
An apparatus and method for applying anti-copy protection to a raster-scanned video signal embodying the invention are described in more detail below with reference to the drawings. In a first aspect, a square waveform is added to the vertical synchronisation pulses of the video signal. The effect of the waveform is not apparent until it has been recorded by a video cassette recorder. The anti-copy protected video signal can therefore be viewed normally on a display device at or near to the picture quality intended by the originator of the video signal, but once recorded, exhibits a reduction in picture quality which makes the video signal quite unattractive to view.
In a second aspect, a positive-going pulse is added to the back porch of the raster scanned video signal such that it is substantially contiguous with the negative-going horizontal synchronisation pulse. The effect of the pulse is again apparent following recording of the video signal by a video cassette recorder. A negative-going pulse 96 may be added after the colour burst 92 and a further positive-going pulse 98 at the start of the active line 84.
A circumvention apparatus and process are also provided to remove the protection from an anti-copy protected signal.
The invention in a preferred embodiment will next be described in detail, by way of example, and with reference to the drawings in which:
The picture signal is comprised of 625 lines 4 containing control information 6, such as synchronisation pulses, which are used to configure the response of the television receiver to the signal and ensure generation of a good quality picture, and usually picture information 8. The line numbers 1-9, 308-320, and 622-625 are shown on the figure.
The synchronisation pulses are used to reset the television receiver so that it is ready to display the next line of picture information, following the completion of the previous line (horizontal synchronisation pulses), or following completion of an entire field (vertical synchronisation pulses).
Following the lines containing picture information shown at the top-left of
In a PAL (or NTSC) signal, the picture is generated on the screen in two rasters that are interlaced with each other, and it is therefore necessary to have two blanking regions. These are both illustrated for the PAL signal in
The first blanking region 12 begins at line 310 of the picture signal and the second blanking region 16 begins at line 623.5. Line 623 of the picture signal therefore contains only a half line of picture information. However, the television receiver is set up such that this line lies outside of the viewable area of the television screen and, like the blanking section, is therefore not viewed. It will be appreciated by those skilled in the art that the lines of the picture signal are numbered consecutively from 1 to 625, and that the signal between the two blanking regions is therefore continuous.
The blanking section itself comprises five negative-going equalisation 18 pulses each of 2.3 μs width, followed by five negative-going vertical synchronisation pulses 14 each of 27.3 μs width, followed by a further five equalisation pulses. The equalisation pulses 18 play a similar role to the horizontal synchronisation pulses 10 and need not be discussed further here as their function is well known to those skilled in the art.
The dc level 20, namely the voltage from which the line synchronisation pulses extend is called the blanking level. This voltage corresponds to the colour black in the picture information.
Following the blanking region there are typically a number of lines 22 of suppressed video information as shown in
As discussed above, the vertical synchronisation pulses control vertical fly-back between fields. A television receiver detects the presence of the vertical synchronisation pulses using an integrator circuit. A capacitor in the integrator circuit charges, during each of the synchronisation pulses, and discharges during the spaces in between. The positioning and the duration of the vertical synchronisation pulses is such that the capacitor charges beyond a predetermined threshold that would not be reached due to charging from the narrower horizontal synchronisation pulses or equalisation pulses alone.
The integrator circuit 30 shown comprises two input terminals 32 and two output terminals 34. As is well know in the art, a resistor 36 is connected between first input and first output terminals, to form the first stage of the circuit, and a capacitor 38 is connected, after the resistor, across the output terminals to form the end-stage of the circuit.
As horizontal or vertical synchronisation pulses are received at the input terminals, the voltage across the capacitor increases as shown in
First Aspect of the Invention
In accordance with the first aspect of the invention, a video signal is modified such that it can be viewed on a television screen in the same way as an unmodified signal, but following recording by a conventional Video Cassette Recorder (VCR), subsequent playback of the recorded signal is unwatchable due to poor picture quality. The reduction in the picture quality in particular appears as instability in the vertical hold of the picture causing it to jump and jitter on the screen.
This effect is achieved by adding an additional waveform into the vertical synchronisation pulses of the signal to interfere with the operation described above with reference to
The top half of
The synchronisation pulses shown in the top half of
The presence of the modulation or additional waveform 52 in the modified signal does not have an effect on vertical synchronisation of the picture until it is recorded. This is because the amplitude of the additional waveform is too small before recording to have any significant effect on vertical synchronisation pulse detection. As a result the modified signal may be viewed normally.
However, when the modified signal is recorded by a video recorder, the signal is amplified. As will be appreciated by those skilled in the art, during recording the magnitude of the part of the signal lying below the blanking signal is ‘clamped’ or limited such that, even after amplification, the amplitude of the negative-going horizontal and vertical synchronisation pulses remains substantially at the synchronisation pulse level, that is the predetermined voltage required to ensure operation of the respective synchronisation pulse detection circuits.
As a result, the signal that is above the blanking level is effectively amplified in relation to the negative part of the signal and therefore takes up a larger proportion of the total amplitude of the signal. In the case of the modified signal, this means that the positive part of the additional waveform is amplified in relation to the negative part.
This is illustrated in
Preferably, the additional waveform is a square wave, as shown, as this is found after amplification to have a greater effect on the zero-crossing point or average value than a substantially sinusoidal waveform for example. The square wave does not necessarily have to have a duty cycle of 50%. Before recording the positive amplitude of the signal might typically extend to +250 mV.
The amplitude of the additional waveform applied to the vertical synchronisation pulse is not sufficient, before recording, to cause instability in the vertical hold of the picture when viewed, either directly on the television or monitor, or indirectly through the video recorder. However, when the signal is recorded, the part of the additional waveform lying above the blanking signal is amplified in relation to the negative-going synchronisation pulse, which remains clamped at the same amplitude. This is illustrated in
Depending on the type of video recorder, an initial modulated signal of amplitude +250 mV might be amplified to +500 mV above the blanking level, while the signal below the blanking level remains clamped at an amplitude of −300 mV. As a result, the average level of the signal in the region of the field synchronisation pulses is now less negative than before it was recorded. The zero-crossing point or average value of the signal in the region of the vertical synchronisation pulse is raised, and the capacitor does not charge as rapidly towards the predetermined detection threshold. The television receiver cannot therefore detect the field synchronisation pulses as easily, and the resulting recorded picture, when played-back, will jump and jitter and not be enjoyable to watch.
Depending on the television receiver, the effect of the amplification of the modulation signal in the field synchronisation pulses might be sufficient after just a single recording to result in the television receiver having difficulty detecting the pulses. Nevertheless, each time the modified video signal is re-recorded the amplitude of the modulation signal is amplified in the positive direction, while the negative amplitude of the synchronisation pulse remains clamped and constant. The disruptive effect of the modulation signal therefore becomes worse each time the modified signal is recorded, so that even if a first unauthorised recording can be made, further recordings of the first unauthorised recording are likely to be unwatchable.
The above description is a simplified explanation of the preferred system. It is not for example necessary for the upper part of the additional waveform to be above the vertical blanking level for the protection effect to be encountered. If for example a square wave is added so that its minimum voltage level is below −300 mV, and its maximum voltage level is below zero, then the average voltage for the additional waveform will be less than −150 mV. If the vertical synchronisation pulse containing the additional waveform is then recorded, the synchronisation pulse and the waveform will be amplified. During recording, the minimum level of the pulse will then be clamped at −300 mV effectively discarding that part of the additional waveform with a lower voltage value. As a result, only the higher part of the additional waveform will remain, thereby raising the average voltage level of the synchronisation pulse.
Furthermore, as most recorders boost the higher frequency part of the signal more on recording than the lower frequency part, each time the modified signal is recorded, the high frequency modulation is amplified more than the vertical synchronisation pulse in which it is disposed. The effect of the modulation is therefore made worse with each successive recording.
Referring again to the upper signal illustrated in
The technique of vertical synchronisation pulse modification described can be used in conjunction with the line synchronisation pulse removal and modulation technique described in International Patent Application WO 01/76240. It has been found that the section of the blanking region in which the line synchronisation pulses have been removed, according to the method disclosed in that patent application, is more frequently mistaken by the television receiver as a vertical synchronisation pulse when the vertical synchronisation pulses are also modified in accordance with this invention such that they are less easily recognised.
The above described techniques for producing an video signal with anti-copy protection have the advantage that the resulting modified signal may be viewed on a television set through the video channel or otherwise, without any disruption to the picture quality. Only when the signal has actually been recorded on a video recorder and is being played-back is the effect on the picture quality realised. This technique has considerable application to pay-per-view broadcasts, in which a broadcaster may, by transmitting a modified picture signal in accordance with the preferred embodiment of the invention, prevent a receiver of the video signals from recording them onto tape. This allows broadcasting companies to broadcast programmes or movies before they are due for release on video, and be sure that any subsequent licensed sales of the programmes or movies on video will be substantially unaffected. As the receiver of the programme cannot record it, the copyright in the programme may be protected.
The vertical synchronisation pulses of both PAL and NTSC signals can be modified in the manner described above.
The set-top-box 60 shown in
The output of the square wave generator 72 is the combination of the square wave and the vertical synchronisation pulse, which form the modified synchronisation pulse shown in
Although a square wave generator has been described with reference to the preferred embodiment, other types of waveform generator could also be employed.
Mixer 68 combines the signal containing the modified vertical synchronisation pulses from the square wave generator 72 and the original signal received from the input 64, such that the unmodified vertical synchronisation pulses of the original signal are replaced by the modified vertical synchronisation pulses. The horizontal synchronisation pulses and the original picture signal remain effectively unaltered in the mixer.
The modified signal is then supplied to an amplifier 74, which amplifies the signal and supplies it to SCART connector 76 or a BNC connector (not shown), or to an RF output terminal 78.
The signal can then be viewed on a television or monitor in the usual way, but is rendered unwatchable when recorded by a VCR.
Although this aspect of the invention has been described with reference to the PAL broadcast television standard, it may also be applied to other broadcast standard signals such as NTSC.
The above described technique cannot be directly used to protect video signals that are to be recorded on video cassettes for distribution. This is because the recording of the original video signal onto any legitimate copies will be subject to the same amplification of the modified vertical synchronisation pulses and any copies will therefore be of a reduced picture quality.
Second Aspect of the Invention
A modified video signal in accordance with the preferred embodiment of the invention, which provides a protection method against video-to-video copying, will next be described with reference to
Video information 84 is represented by the stepped waveform ascending from the blanking or black level at its bottom to the white peak level at its top. It will be appreciated that this waveform if viewed on a television screen would be seen as colour bars.
Between the active lines are negative-going horizontal synchronisation pulses 86 which control flyback in the television receiver between successive lines of the video signal. The horizontal region on the left of a horizontal synchronisation pulse 36 is known as the front porch 88, while that on the right is known as the back porch 90. Colour burst information 92, used by the television receiver during demodulation of the chrominance part of the video signal is located on the back porch of the signal on each line and is represented purely diagrammatically by a rhombus or diamond shape.
The video signal in
The modified PAL signal includes a first additional pulse 94 with a magnitude that is approximately equal to the peak-white level; typically at a level of 1 to 1.2 volts. The first additional pulse is substantially contiguous with the horizontal synchronisation pulse. As shown it is situated directly adjacent to the horizontal line synchronisation pulse. In fact the ascending or right-hand edge of the horizontal synchronisation pulse meets the ascending or left-hand edge of the additional pulse smoothly to form a continuous slope, that is to say that there is substantially no space between the edges of the two pulses.
The presence of the pulse at this location has been found to have an adverse effect on the Automatic Gain Control (AGC) circuits of video recorders. These circuits detect horizontal synchronisation pulses and, based on a determination of their amplitude, amplify the video signal such that it is suitable for recording. The combination of the negative-going horizontal synchronisation pulse 36 and the positive-going additional pulse 94 appears to the automatic gain control circuit as a larger than normal horizontal synchronisation pulse. As a result the amplification provided by the automatic gain control circuit is less than it should be and is insufficient in respect of the rest of the video signal. The recorded signal, when played back later on a television receiver, will be too dark to be satisfactorily viewed. Furthermore it may exhibit stability problems, as the horizontal synchronisation pulses and vertical synchronisation pulses have not been amplified enough to be reliably detected by the circuits of the television receiver.
As mentioned above, it is important for the implementation of this feature that the additional pulse be placed contiguous with the horizontal synchronisation pulse. In particular, the pulse is not placed in the colour burst part of the signal, as this has been found to have an adverse effect on the picture quality while having little or no effect on the automatic gain control circuits.
The amplitude of the pulse 94 is approximately white level, as shown, though it may be that an amplitude above 30% of white level will be sufficient in certain circumstances. The duration of the pulse 94 is between 0.5 and 2 μs and is preferably about 1 μs. If the duration of the pulse 94 is at the larger end of the range, the horizontal synchronising pulse 86 can be moved slightly earlier (to the left) to provide sufficient space.
The presence of the first pulse 94 has however been found to have a detrimental effect on the playability of the modified signal even before unauthorised recording has occurred. This is because the pulse interferes with the detection stage circuits in the television receiver causing the receiver to display the modified signal as a picture that is darker than it should be. In order to compensate for this effect, a second, negative going pulse 96, substantially equal in magnitude to a normal horizontal synchronisation pulse, is preferably added to the signal directly after the colour burst. This additional negative-going pulse has been found to reverse the ill-effect of first pulse 94 on legitimate playback of the original modified signal. The duration of pulse 96 is in the range 1 to 5 μs, and is preferably 1.8 μs.
The presence of second pulse 96 however also makes possible the addition of a third, positive-going pulse 98, located just prior to or at the start of the active video information contained in the signal, and directly next to the negative-going pulse 96. The magnitude of the third pulse is about the same as that of the peak white level, and its duration is in the range 1 to 4 μs, preferably 2 μs.
The third pulse acts in the same way as first pulse 94, by interfering with the operation of the automatic gain control of the video recorder. The combination of the pulses 96 and 98 again appear like a horizontal synchronisation pulse of larger magnitude than an ordinary synchronisation pulse, and therefore interferes with the automatic gain control circuits of the video recorder to add to the effect caused by the first pulse 96. Without second pulse 96, the presence of third pulse 98 is not enough to cause sufficient instability in the recorded signal.
The combination of first pulse 94 and second pulse 99, shown in
The two different signals illustrated in
The preferred apparatus for adding pulses 94, 96 and 98 to the video signal is shown in
The apparatus 100 comprises a housing 102 in which an external input 104 is mounted for receiving an unmodified video signal. The signal is passed from the input to a digital video processor 106 controlled by control software 108. The digital processor analyses the received unmodified signal, and under the control of the software 108, adds pulses to the signal in the manner described above. The digital video processor outputs a signal containing the original signal and the additional pulses to video amplifier 110. This amplifies the signal for output to a video output 112 external to the apparatus. The video output thereby supplies a video signal that has been protected against unauthorised copying according to the second aspect of the invention described above. The video distributor can then record this onto a video cassette using a professional video recorder in which the automatic gain control circuit is turned off.
In addition to the protection methods described above, we have appreciated that it is also desirable to provide a system and method of defeating or circumventing such protection. Thus, a protected signal could be processed to remove the protection and give an unprotected signal that could be copied once again. A circumvention system 120 for doing this is described in more detail in
A protected signal is first received at input 122. It is then passed to switch 124 and to Horizontal and Vertical Synchronisation Pulse detection circuit 126 respectively. The Synchronisation Pulse detection circuit 126 detects the presence of a vertical or a horizontal sync pulse, and sends a signal to controller 128. The presence of the modifications in the region of the vertical synchronisation pulse should not affect the ability of the detection circuit 126, as at this stage they will not have been recorded. In any case, the time constant of the detection circuit 126 in the circumvention system is preferably smaller than that of the corresponding circuitry in a television receiver. In this way, it may be possible to remove the vertical synchronisation pulse protection from signals which have been copied. The LM1881 chip may for example be used as the Horizontal and Vertical Synchronisation Pulse detector 126, with the SX20 chip used for the controller.
Depending on the output from the synchronisation pulse detection circuit 126, the controller 128 operates the switch 124 to swap between first and a second input. If no synchronisation pulse is detected, the controller 128 selects input 122 supplying the protected signal. As will be recalled from the above discussion, the part of the protected signal away from the synchronisation pulses is not modified in the protection process and can be supplied to the output directly.
If a horizontal synchronisation pulse, or a vertical synchronisation pulse is detected however, the controller 128 selects the second input of the switch, which is connected to signal generator 130. The signal generator 130 provides a signal to overwrite or replace that part of the signal input at 122, which contains the modifications so that the protection is no longer present. In the case of the modified vertical synchronisation pulse, for example, the signal generator 130 preferably provides an unmodified vertical synchronisation pulse, whereas in the case of the horizontal synchronisation pulse, the signal generator preferably provides an unmodified back porch. The timing applied to the switch operation is therefore precisely controlled to ensure that only the previously modified areas of the signal are replaced. For this reason, the signal generator receives an input from the controller 128 ensuring that it is synchronised to the horizontal or vertical synchronisation pulses detected by detector 126. In the case of the modifications to the horizontal line synchronisation pulse, this may mean swapping the switch back to the input signal at the appropriate time to maintain the colour burst in the signal sent to the output.
As mentioned above, the signal generator may simply overwrite the protection modified areas of the signal. In this case, the signal generator 130 may comprise a circuit that gives a controllable, constant output voltage, or a circuit that simply replaces the affected area of the signal with an unaffected area. In this way the vertical synchronisation pulse, or the back porch region may simply be replaced. On the other hand, the removal of the modified areas may comprise addition of a pulse or wave form to cancel that already in place in the signal. In such a case, the signal generator may comprise a square wave generator or pulse generator.
Thus, reconstructing the unprotected signal may involving switching the signal back to its original level at the appropriate time, replacing the modified signal with an unmodified signal, or adding a compensating signal.
The output from switch 124 is then passed to amplifier 132, and finally to output 134. The AD8051 chip may be used as the amplifier.
Although the preferred apparatus and circumvention device have been described with reference to a digital video processor and control software, it will be appreciated that both the control software and the digital control software could be implemented in purpose-built equivalent electronic circuits. Also, although reference has been made throughout this application to a television receiver, it will be understood that any display device on which video signals can be viewed and which operates in a way equivalent to a television receiver to display a raster-scanned signal is included within the scope of the invention. Furthermore, the operation of any apparatus described in this application could be implemented partly or wholly in software as appropriate.
As is known in the art, both positive and negative transmission schemes exist, so called because of the polarity of the signal carrying part of the signal. The schemes described in the application above, by way of example, are positive transmission schemes, and the orientation of the additional pulses as ‘positive’ or ‘negative’ reflect this terminology. It will however be appreciated that the invention also has application to negative transmission schemes. In this case, positive going pulses will be negative going pulses, and negative going pulses will be positive going pulses.