GB2116393A - Colour television signal processing - Google Patents

Abstract

In apparatus for processing 525- line N.T.S.C. colour television signals, in which in conjunction with each line of the signal there is provided an additional signal for use therewith, the additional signal is formed essentially of the luminance component of a signal equivalent to the signal M 525 lines previous to the current line, and the chrominance component of a signal substantially equivalent to the signal T 525 or B 526 line previous to the current line, or an average of these, the form of the chrominance component being selected automatically by a circuit 80 in dependence upon correlation of vertical chrominance detail in lines B, M and T. The circuit 80 for determining vertical detail, which is in principle also applicable to PAL signals. provides moduli DB and DT of the differences between the chrominance components of lines B, M and M, T, respectively and produces therefrom a signal: <IMAGE> determining whether, and in what sense, a signal 1 DIVIDED 2 (T-B) from subtractor 64 is combined with a signal 1 DIVIDED 2(T+B-2M) from adder 62. The resulting signal is filtered for the chrominance component and added to the full line M at 72. <IMAGE>

Description

SPECIFICATION Colour television signal processing This invention concerns apparatus for determining vertical detail in a colour television signal.

This application is divided out of our application No.80 12449 (Serial No.2054313).

The present invention is defined in the appended claim to which reference should now be made.

The invention will be described by way of exampie with reference to the drawings, in which: Figure 1 shows a block circuit diagram of circuit for processing 525-line N.T.S.C. colour television signals which inciudes a circuit for determining vertical detail in accordance with the invention, and Figure 2 shows a possible re-arrangement of the circuit of Figure 1.

In Figure 1, an input 50 is connected to a 2N line (e.g. 524 line) delay 52, which is in turn connected to two one-line delays 54, 56 arranged in series. The signal outputs of the three delays are denoted by B, M and T, designating the bottom, middle and top of three successive lines.

Two subtractors 58, 60 provide respectively the signals T-M and B-M subtraction across the oneline delays. A halving adder 62 adds the outputs of the two subtractors 58, 60 and a halving subtractor 64 subtracts them. The output of subtractor 64 is multiplied by a factor a in a multiplier 66, and the resultant added in an adder 68 to the output of adder 62. A chrominance band-pass filter 70 receives the output of adder 68 and adds it in an adder 72 to the output M of delay 54 after the latter has passed it through an equalising delay 74. The circuit output is provided on a line 76.

By varying the value of the factor a the circuit of Figure 1 can be made to operate in different ways. Thus if a=O, the output comprises over the chrominance frequency band a signal equivalent to the average of the signals 524 lines and 526 lines previous to the current line. If a=+1 the chrominance component is based on the signal 524 lines or 526 lines previous to the current line.

In all cases, over the luminance band, ie the band below the chrominance band, the output is based on the signal 525 lines previous to the current line.

The value of a is controlled by a control circuit 80 in response to the signal itself. The control circuit detects vertical chrominance detail by comparing the differences between the chrominance signal in the reference line (M) and the two neighbouring lines (B, T,). For this purpose, the control circuit 80 includes two adders 81 which add the signal M from delay 54 to the signals T and B respectively, these being the output of delays 56 and 52. Chrominance bandpass filters 82 are connected to the outputs of the two adders 81 respectively, and the output of each filter 82 is rectified at 84. Low-pass filters 86 are connected respectively to the outputs of modulus circuits 84. The outputs of the low-pass filters are subtracted in a subtractor 87 and added in an adder 88, and the output of the subtractor is divided by the output of the adder in a divider circuit 90.The output of the divider constitutes the factor a. The circuit 80 is in principle applicable to PAL signals, with appropriate alteration to the elements 81.

Because in the N.T.S.C. system the subcarrier frequency is essentially equal to an odd integral multiple of half the line frequency, there is an effective subcarrier phase reversal between alternate lines. In order to subtract the chrominance components of the signal, it is therefore necessary to add the signals themselves and this is done in adders 81. The outputs of the chrominance band-pass filters82 thus represent respectively the difference between the chrominance components of the top and middle of a set of three lines, and the difference between the bottom and middle lines. The absolute magnitudes of these differences are taken, and the low-pass filters smooth the instantaneous rectified chrominance signals to give a measure of the amplitude of the chrominance differences.

These differences can be termed: DfflCTCM and DB=CB CM where C is the chrominance vector composed of I and Q components.

The circuit components 87, 88 and 90 then operate on the signals DT and D6 to give a, in accordance with the relation: (DB-DT) a= (DB+DT) If the reference line content is closer to the upper line, then more weight is given to the upper prediction, i.e. a > 0, and vice versa if the reference line is closer to the lower line.

As the detector 80 must operate on band-pass signals, it is more economical to rearrange the circuit as shown in Figure 2. The detector 80 does not then need to include the chrominance bandpass filters 82, but an additional one-line delay 92 is required. This is preferable in particular for digital processing.

When a digital realization is used, it is preferred for the sampling structure to be line locked.

The output signal can, as described in our German Patent Application P 28 16 236 and British Patent Application 15810/77 (Patent 1,588,193), be used for prediction in noise reduction, for temporal DPCM, for error concealment or for frame freezing.

Claim

1. Apparatus for determining vertical detail in a colour television signal, comprising means for deriving signals representing the differences

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (1)

1. **WARNING** start of CLMS field may overlap end of DESC **.

   SPECIFICATION Colour television signal processing This invention concerns apparatus for determining vertical detail in a colour television signal.

   This application is divided out of our application No.80 12449 (Serial No.2054313).

   The present invention is defined in the appended claim to which reference should now be made.

   The invention will be described by way of exampie with reference to the drawings, in which: Figure 1 shows a block circuit diagram of circuit for processing 525-line N.T.S.C. colour television signals which inciudes a circuit for determining vertical detail in accordance with the invention, and Figure 2 shows a possible re-arrangement of the circuit of Figure 1.

   In Figure 1, an input 50 is connected to a 2N line (e.g. 524 line) delay 52, which is in turn connected to two one-line delays 54, 56 arranged in series. The signal outputs of the three delays are denoted by B, M and T, designating the bottom, middle and top of three successive lines.

   Two subtractors 58, 60 provide respectively the signals T-M and B-M subtraction across the oneline delays. A halving adder 62 adds the outputs of the two subtractors 58, 60 and a halving subtractor 64 subtracts them. The output of subtractor 64 is multiplied by a factor a in a multiplier 66, and the resultant added in an adder 68 to the output of adder 62. A chrominance band-pass filter 70 receives the output of adder 68 and adds it in an adder 72 to the output M of delay 54 after the latter has passed it through an equalising delay 74. The circuit output is provided on a line 76.

   By varying the value of the factor a the circuit of Figure 1 can be made to operate in different ways. Thus if a=O, the output comprises over the chrominance frequency band a signal equivalent to the average of the signals 524 lines and 526 lines previous to the current line. If a=+1 the chrominance component is based on the signal 524 lines or 526 lines previous to the current line.

   In all cases, over the luminance band, ie the band below the chrominance band, the output is based on the signal 525 lines previous to the current line.

   The value of a is controlled by a control circuit 80 in response to the signal itself. The control circuit detects vertical chrominance detail by comparing the differences between the chrominance signal in the reference line (M) and the two neighbouring lines (B, T,). For this purpose, the control circuit 80 includes two adders 81 which add the signal M from delay 54 to the signals T and B respectively, these being the output of delays 56 and 52. Chrominance bandpass filters 82 are connected to the outputs of the two adders 81 respectively, and the output of each filter 82 is rectified at 84. Low-pass filters 86 are connected respectively to the outputs of modulus circuits 84. The outputs of the low-pass filters are subtracted in a subtractor 87 and added in an adder 88, and the output of the subtractor is divided by the output of the adder in a divider circuit 90.The output of the divider constitutes the factor a. The circuit 80 is in principle applicable to PAL signals, with appropriate alteration to the elements 81.

   Because in the N.T.S.C. system the subcarrier frequency is essentially equal to an odd integral multiple of half the line frequency, there is an effective subcarrier phase reversal between alternate lines. In order to subtract the chrominance components of the signal, it is therefore necessary to add the signals themselves and this is done in adders 81. The outputs of the chrominance band-pass filters82 thus represent respectively the difference between the chrominance components of the top and middle of a set of three lines, and the difference between the bottom and middle lines. The absolute magnitudes of these differences are taken, and the low-pass filters smooth the instantaneous rectified chrominance signals to give a measure of the amplitude of the chrominance differences.

   These differences can be termed: DfflCTCM and DB=CB CM where C is the chrominance vector composed of I and Q components.

   The circuit components 87, 88 and 90 then operate on the signals DT and D6 to give a, in accordance with the relation: (DB-DT) a= (DB+DT) If the reference line content is closer to the upper line, then more weight is given to the upper prediction, i.e. a > 0, and vice versa if the reference line is closer to the lower line.

   As the detector 80 must operate on band-pass signals, it is more economical to rearrange the circuit as shown in Figure 2. The detector 80 does not then need to include the chrominance bandpass filters 82, but an additional one-line delay 92 is required. This is preferable in particular for digital processing.

   When a digital realization is used, it is preferred for the sampling structure to be line locked.

   The output signal can, as described in our German Patent Application P 28 16 236 and British Patent Application 15810/77 (Patent 1,588,193), be used for prediction in noise reduction, for temporal DPCM, for error concealment or for frame freezing.

   Claim

   1. Apparatus for determining vertical detail in a colour television signal, comprising means for deriving signals representing the differences between the chrominance components of two pairs of lines of the signal, means for taking the modulus and low-pass filtering each of these signals, and means for evaluating the resultant of the difference divided by the sum of the filtered signals.