GB2173668A - Improvements in or relating to NTSC processing systems - Google Patents

Abstract

A system for decoding NTSC colour video signals in which three adjacent lines of colour video signals are provided by two line delays 1,2. An analyser circuit (Figure 1) produces signals representing the positions of any chrominance change. These signals are used to set values K1, K2 which control the decoding of the luminance signals so that the high frequency part is obtained either from the upper two lines, or the lower two lines or the middle line alone depending on the position of the chrominance change. <IMAGE>

Description

SPECIFICATION Improvements in or relating to NTSC processing systems This invention relates two improvements in NTSC Processing Systems particularly to those involving signal decoders.

The use of adaptive colourtelevision signal decoders in television broadcast studies is well known. They may be used for example when signals are received in one standard, say NTSC and it is required to convert them to another standard, or when editing is required which has to be carried out on the colour component signals rather than the encoded signals. In performing such adaptive decoding the signalsto be decoded are analysed in a number of ways in a so called error detector e.g. by correlating successive lines of colour signals and by devising a control signal which is applied to vary the decoding process.For example the control may depend on the degree of difference existing between the chrominance in adjacent lines ofthe colourwave form, and it may be used to selectfordecoding ata particulartimethe luminance signal from one or other ofa number of adjacent lines, which maydifferfrom the linefrom which the chrominance signal is decoded.

The present invention has for its object two provide an improved decoder ofthe general kind and according to the invention the analyser circuit is arranged to produce a plurality of control signals which respectively selectfordecoding luminance signals from one of a numberoflines orthe average of luminance signalsfrom a number of such lines.

One embodiment of the present invention will now be described with reference to the following diagrams, Figure 1 shows one example of the error detector which forms part of the invention.

Figure2shows one example of the NTSC Decoding system.

The analyser circuit in Figure 1 detects differences in chrominance signals between adjacent lines of video signals and provides signals to the decoder to control the decoding dependent on errors. Incoming lines of colour video signals in NTSC format are su bjectto line delays on 1 and 2 thatthree adjacent lines ofvideo signals are available for decoding.

NTSC signals are colour signals in which the chrominance sub-carrierfrequency is such that there is effectively a sub-carrier phase reversal on alternate lines. The incoming video signals are composed of a luminance signal, a chrominance signal and signal reference information and the luminance and chrominance signal must be separated to produce a colour image on a screen.

In the analyserthe video signals on path 4 are delayed by one line period and those on path 5 by two line periods and in this way three adjacent lines of video signals are provided for error detecting. An error is implied if there is a change in the sub-carrier amplitude or phase as between adjacent lines, though such a change may represent vertical chrominance detail. The three lines of video signals on paths 3,4 and 5 are passed through band pass filter 6,7 and 8 to produce signals only ofthe high frequency luminance and the chrominance components ofthe signal. As the sub-carrier signal phase is reversed on alternate lines then, if there is no error, summing the video signals on path 3 and 4and 4 and 5 should produce high frequency luminance components only.This summing is performed in adders 9 and 10 and the output of 9 should bethe average of the high frequency luminance components on paths 3 and 4 and the output of 10 should be the same component but for paths 4 and 5. If, however, there is a change in chrominance in one ofthe linesthen there will be an additional partofsignal outputfrom the adders. The outputs ofthe adders9 and 10 arefull wave rectified in 11 and 12 to produce the absolute amplitudes and these are then subtracted in 13. The output of 13 gives an indication of the errors in the lines but at this stage cannot distinguish where the error is.If the error is between signals in paths 3 and 4 and there is no error between 4 and 5 the output of 13 will bethe difference in high frequency luminance components betweenthetop and bottom lines, which will be negligible, andthe negative absolute amplitude of the error between paths 3 and 4. In the case of an error between paths 4 and 5 the output will be the same except the amplitude will be positive. However, if there is say a small error between paths3 and 4 and a large error between 4and Sthiswill only appearas an error between 4and 5.To obtain information on the error at the centre line, path 4, a second difference is taken in subtractor 16. The signals from band pass filters 7 and 8 arefullwave rectified in 14 and 15 and the difference taken in 16. This gives information on the errors between signals on paths4 and 5.The outputs from the subtractors are passed through low pass filters 17 and 18to remove spurious sub-carrierfrequency signals. Logic circuit 19 produces a signal K1 in response to the output ofsubtractors 13 and 16 and the signal is passed to the decoder to control the decoding. The signal K2 is provided by subtractor 16 and passed through error stretching circuit 20 and is related to the error in the middle line. The signals K1 and K2 control which of the three lines of video signals are to be used by the decoder to produce the output liminance signal. If output from the subtractor indicates an error in the lowertwo lines, or as in the case discussed where the error is greater between the lower two lines logic 19 outputs K1 = 1.For an error between the upper two lines K1 = -1 and for no error(orerrors that cancel) K1 = O. K2takes a value 1/2 unless K, = -1 which indicates an error in the upper two lines if a difference is detected in the lowertwo lines also all the lines must be different and K2 goesto zero.

The decoder shown in Figure 2 also receives three adjacent lines of video signals delayed by delays 1 and 2 and these are available on paths 21,22 and 23. It is to be understood that in this and in the error detecting circuit all circuit components are assumed to be delayless. In practice corresponding delays are introduced as required. The signals on paths 21,22 and 23 are passed through band pass filters 24,25 and 26to produce signals composed ofthe high frequency luminance component and chrominance components only. The signals from band pass filters 24 and 25 are passed through the divide by two circuits 37 - 40 and added in adde 27 to produce a signal composed ofthe average ofthe high frequency luminance components on paths 21 and 22, this is marked as HFU. In this case the chrominance components are assumed to cancel.Adder28 produces an output of the average ofthe high frequency luminance components on paths 22 and 23, marked as HF. In multiplier 31 the outputofsubtractor is is multiplied bythefactor K1 from the error detector and then the output of the multiplier is added in 32to the output of adder 30. K1 is limited to the values 0,1 or-1.

The signal which is outputfrom adder 32 is then: a = K1 (HFu-HFL)+(HFL+HFu) IfK1 = O a = HFU + HFL IfK1 = 1 a=2HFuand IfK1 = -1 a = 2HFL Signal a is multiplied in multiplier 33 by signal K2 and then subtracted from the output of band pass filter 25.

This output consists ofthe high frequency luminance and the chrominance components of the middle line HFM and CHM. The output b ofthe subtractor 34 is b = K2a + CHM + HFM, where K2 can take the values 0,1/2.

Whateverthe values of K1 and K2, the chrominance component of the signal b is composed mainly ofthe chrominance signal from the middle line and this is used as the separated chrominance signal which is ultimately decoded.

The signal b is as follows: IfK2 = 0 b=CHM+HFM K2= 1/2 b = HFM + CHM-(HFU + HFL) 2 In both cases K1 =0 BUTHFU = HF/2 (from path 21) + 2HFM AND HFL = HF (from path 23) + HFM 2 2 Therefore forthe second case above, b = CHM -(HF(21) + HF(23)) + HFM 4 2 It is assumed that the high frequency luminance components are the same in the three lines b= CHM.

Similarly IfK2=1/2 b = CHM + HFM-HFU and K1 +1 lfK2=1/2 b = CHM + HFM-HFL K1=-1 Therefore b = CHM in all but the case when K2 = 0.

When K2 = 0 the decoder acts in the simple mode and the chrominance component includes the high frequency luminance component from the middle line. To obtain the luminance signal the signal b =k2a + CHm + HFm is subtracted from the signal on path 36 in the subtractor35. The signal on path 36 consists ofthe video signal from the middle line timed to coincide with the output of band pass filter 25. The coutputof subtractor35 is LM + HFM + CHM - b = Lm - K2a, so depending on the values of K2 and K1 available from the error detector the luminance signal will include the high frequency component from different lines or different combination of lines.

If K2 = Othen all three lines are different and the output of the subtractorwill be Lm only and the decoderwill be working in the so called simple mode. If K2 = 1/2 K = 1 output will be Lm + HFU similarly K = -1 Y out will be Lm + HFLorK=OY= Lm + (HFL + HFU) which is the average of all three lines.

Claims (7)

1. A decoder for NTSC colour signals comprising, an analyser circuit arranged to produce a plurality of control signals, representing the difference in chrominance between incoming lines of video signals, means for decoding luminance signalsfrom oneofa number of lines ortheaverage ofsignalsfrom a number of such lines.

2. An analyser circuit for NTSC colour signals comprising, means for providing three adjacent lines of video signals, means for detecting a difference in chrominance between said lines, and means for providing control signals dependent on said difference, said control signals being dependent on the position of detected differences.

3. An analysercircuit as in claim 2wherein said detecting means includes means for producing a signal representing the difference between the absolute values of the sum of two adjacent lines and the sum of one of said two lines with the next adjacent line, and means for producing a second signal representing the difference between the absolute values of two adjacent lines, said difference signals being used to produce the control signals.

4. An analyser as in claim 3 comprising means to produce two control signals, one control signal having three values dependent on the position of the error between the three lines, the second signal having one value when there is an error between the two lines and a second value when no error is present.

5. Adecodercircuitfor NTSC systems comprising, means for providing three adjacent lines of video signals, analyser means for providing control signals representing the position of a change in chrominance, means dependent on said control signal for decoding the high frequency luminance part of the video signals, said means being such that in one mode the signals are decoded from thefirst two lines of signals, in the second mode from the othertwo lines of signals and in the third mode from the middle line of signals alone, and in the fourth modefromthe average ofallthreelines.

6. A decoder as in claim 5 wherein said means for decoding operates in the first mode when the control signals indicate a change in chrominance between the second two lines, in the second mode when the change is between the firsttwo lines and in the third mode where there is no change and in thefourth mode when all three lines aredifferent.

7. A decoder substantially as described herein with reference to the accompanying diagram.