GB2073535A

GB2073535A - Colour television transcoding matrices

Info

Publication number: GB2073535A
Application number: GB8011205A
Authority: GB
Original assignee: British Broadcasting Corp
Current assignee: British Broadcasting Corp
Priority date: 1980-04-03
Filing date: 1980-04-03
Publication date: 1981-10-14
Also published as: GB2073535B

Abstract

Transcoding from R, G, B to (R-Y), Y, (B-Y) is achieved (Figure 1) by forming (R-G) and (B-G), multiplying these by 0.299 and 0.114 respectively, adding the resultants together and to G such as to give Y, and subtracting Y from each of R and B. Gain circuitry may be connected to convert the (B-Y) and (R-Y) signals into U and V signals. Transcoding in the converse direction is achieved (Figure 2) by adding (48) (R-Y) and Y to give R, adding (50) (B-Y) and Y to give B, multiplying (52) (R-Y) by a factor 0.299/0.587, multiplying (54) (B-Y) by a factor 0.114/0.587, adding (56) the outputs of the two multipliers and subtracting (58) the resultant from Y so as to give G. In a circuit receiving Y, U and V signals, a gain adjusting circuit is included before the input to each of adders 48, 50, and the multiplication factors of multipliers 52, 54 are correspondingly adjusted. A minimum of multipliers is required in each case, and multiplier errors do not affect grey scale colours. <IMAGE>

Description

SPECIFICATION Colour television transcoding matrices This invention relates to a transcoding matrix for use with colour television signals. The invention is particularly applicable to transcoding between separated PAL colour television signal components in, on the one hand, R,G,B form and Y,R-Y, B-Y form, on the other hand.

In this specification the letters R,G and B are used to represent the red, green and blue signal components, while the letter Y is used to represent the luminance signal, in the conventional way.

The invention will be described by way of example with reference to the accompanying drawing, in which: Figure 1 is a block diagram of an R,G,B to Y,R-Y,B-Ytranscoding matrix; and Figure 2 is a block diagram of a converse transcoding matrix.

The transcoding matrix 10 shown in Figure 1 includes three inputs 12, 14 and 16 for receiving the R, G and B signals respectively. A subtractor 18 is connected to subtract the green signal from the red signal, and a subtractor20 is connected to subtract the green signel from the blue signal. A multiplier 22 multiplies the output of subtractor 18 by the factor 0.299, and a multiplier 24 multiplies the output of subtractor 20 by the factor 0.114. An adder 26 adds together the green signal at input 14, the output of multiplier 22, and the output of multiplier 24. The output of adder 22 constitutes the luminance or Y output of the matrix. A subtractor 28 receives the R input signal from terminal 12, and subtracts from this the output of adder 26.Similarly, a subtractor 30 receives the blue input signal from terminal 16 and subtracts from this the output of adder 26. The two outputs of the subtractors 28 and 30 thus constitute respectively the R-Y and B-Y output signals.

Each of these two output signals can be multiplied by a respective gain factor to produce respectively the colour difference signals U and V.

It will be seen that in the matrix of Figure 1, the subtractors 18 and 20 respectively provide outputs representative of R-G and B-G. Thus adder 26 forms the following sum: G + 0.299 (R-G) + 0.114 (B-G) This expression can be written as 0.587 G + 0.299 R + 0.114 B which is by definition the luminance signal Y.

The matrix of Figure 1 has particular application where the signals being processed are in digital form. It will be seen that it uses a minimum of multipliers, which can introduce rounding errors.

Figure 2 shows a transcoding matrix for use in the opposite direction. The matrix 40 of Figure 2 includes three inputs 42,44 and 46 for receiving respectively the Y, R-Y, and B-Y signals. An adder 48 adds the Y and R-Y signals to provide an R output signal. An adder 50 adds the Y and B-Y signals to produce an output signal B.

The R-Y signal at input 44 is also multiplied in a multiplier 52 by a factor equivalent to 0.299/0.587.

The B-Y signal at input 46 is similarly multiplied in the multiplier 54 by a factor equivalent to 0.114/0.587. An adder 56 adds the outputs of multiplier 52 and 54, and a subtractor 58 subtracts the output of adder 56 from the Y input signal at input 42. It will be appreciated that the adder 56 and subtractor 58 can be combined into a single combining unit.

The output of subtractor 58 constitutes the G output signal.

In Figure 2 it can be seen that the output signal is equivalent to: Y-[(0.299/0.587) (R-Y) + (0.114/0.587) (B-Y)J From the definition of the luminance signal Y, this it can be seen reduces to the green signal G.

Again a minimum of multipliers are used in Figure 2 which makes this transcoding matrix of particular utility with digital signals.

The matrix of Figure 2 can be adapted to receive Y, U, V signals instead of Y, R-Y, B-Y signals. While this could simply be done by adjusting the gain factor at the inputs 44 and 46, it is preferred to include a gain adjusting circuit immediately before the inputs to adders 48 and 50, as indicated by X on the figure, and to adjust the multiplication factors of multipliers 42 and 44 in accordance with the same gain factors.

In this way the signal reaching the adder 56 are subjected to only one multiplication operation, which reduces the rounding errors caused in a digital signal.

The transcoding matrices of both Figure 1 and Figure 2 have the further especial advantage that any inaccuracies or rounding errors resulting from the operation of the multipliers 22,24 in Figure 1 or 52, 54 in Figure 2 are of no relevance to colours on the grey scale. In Figure 1, for monochrome signals, R, G and B are all equal, so that the output of subtractors 18 and 20 are 0. Similarly, in Figure 2 the inputs to the multipliers 52 and 54 will be 0. This avoids the need to normalise or equalise the signals to remove colour casts on the grey scale.

1. A transcoding matrix for transcoding PALcolour television signals in R,G,B form to R - Y, Y, B - Y form, comprising: first, second and third inputs for receiving R, G and B signals respectively; first combining means connected to the inputs for providing an R-G signal art a first outputthereof and a B-G signal at a second output thereof; first multiplying means connected to the first output of the first combining means for multiplying the R-G signal by a factor effectively substantially equal to 0.299; second multiplying means connected to the second output of the first combining means for multiplying the B-G signal by a factor effectively substan tially equal to 0.114; ; and second combining means connected to the outputs of the first and second multiplying means and the second input to provide a luminance signal Y =

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

Claims

**WARNING** start of CLMS field may overlap end of DESC **. SPECIFICATION Colour television transcoding matrices This invention relates to a transcoding matrix for use with colour television signals. The invention is particularly applicable to transcoding between separated PAL colour television signal components in, on the one hand, R,G,B form and Y,R-Y, B-Y form, on the other hand. In this specification the letters R,G and B are used to represent the red, green and blue signal components, while the letter Y is used to represent the luminance signal, in the conventional way. The invention will be described by way of example with reference to the accompanying drawing, in which: Figure 1 is a block diagram of an R,G,B to Y,R-Y,B-Ytranscoding matrix; and Figure 2 is a block diagram of a converse transcoding matrix. The transcoding matrix 10 shown in Figure 1 includes three inputs 12, 14 and 16 for receiving the R, G and B signals respectively. A subtractor 18 is connected to subtract the green signal from the red signal, and a subtractor20 is connected to subtract the green signel from the blue signal. A multiplier 22 multiplies the output of subtractor 18 by the factor 0.299, and a multiplier 24 multiplies the output of subtractor 20 by the factor 0.114. An adder 26 adds together the green signal at input 14, the output of multiplier 22, and the output of multiplier 24. The output of adder 22 constitutes the luminance or Y output of the matrix. A subtractor 28 receives the R input signal from terminal 12, and subtracts from this the output of adder 26.Similarly, a subtractor 30 receives the blue input signal from terminal 16 and subtracts from this the output of adder 26. The two outputs of the subtractors 28 and 30 thus constitute respectively the R-Y and B-Y output signals. Each of these two output signals can be multiplied by a respective gain factor to produce respectively the colour difference signals U and V. It will be seen that in the matrix of Figure 1, the subtractors 18 and 20 respectively provide outputs representative of R-G and B-G. Thus adder 26 forms the following sum: G + 0.299 (R-G) + 0.114 (B-G) This expression can be written as 0.587 G + 0.299 R + 0.114 B which is by definition the luminance signal Y. The matrix of Figure 1 has particular application where the signals being processed are in digital form. It will be seen that it uses a minimum of multipliers, which can introduce rounding errors. Figure 2 shows a transcoding matrix for use in the opposite direction. The matrix 40 of Figure 2 includes three inputs 42,44 and 46 for receiving respectively the Y, R-Y, and B-Y signals. An adder 48 adds the Y and R-Y signals to provide an R output signal. An adder 50 adds the Y and B-Y signals to produce an output signal B. The R-Y signal at input 44 is also multiplied in a multiplier 52 by a factor equivalent to 0.299/0.587. The B-Y signal at input 46 is similarly multiplied in the multiplier 54 by a factor equivalent to 0.114/0.587. An adder 56 adds the outputs of multiplier 52 and 54, and a subtractor 58 subtracts the output of adder 56 from the Y input signal at input 42. It will be appreciated that the adder 56 and subtractor 58 can be combined into a single combining unit. The output of subtractor 58 constitutes the G output signal. In Figure 2 it can be seen that the output signal is equivalent to: Y-[(0.299/0.587) (R-Y) + (0.114/0.587) (B-Y)J From the definition of the luminance signal Y, this it can be seen reduces to the green signal G. Again a minimum of multipliers are used in Figure 2 which makes this transcoding matrix of particular utility with digital signals. The matrix of Figure 2 can be adapted to receive Y, U, V signals instead of Y, R-Y, B-Y signals. While this could simply be done by adjusting the gain factor at the inputs 44 and 46, it is preferred to include a gain adjusting circuit immediately before the inputs to adders 48 and 50, as indicated by X on the figure, and to adjust the multiplication factors of multipliers 42 and 44 in accordance with the same gain factors. In this way the signal reaching the adder 56 are subjected to only one multiplication operation, which reduces the rounding errors caused in a digital signal. The transcoding matrices of both Figure 1 and Figure 2 have the further especial advantage that any inaccuracies or rounding errors resulting from the operation of the multipliers 22,24 in Figure 1 or 52, 54 in Figure 2 are of no relevance to colours on the grey scale. In Figure 1, for monochrome signals, R, G and B are all equal, so that the output of subtractors 18 and 20 are 0. Similarly, in Figure 2 the inputs to the multipliers 52 and 54 will be 0. This avoids the need to normalise or equalise the signals to remove colour casts on the grey scale. CLAIMS

1. A transcoding matrix for transcoding PALcolour television signals in R,G,B form to R - Y, Y, B - Y form, comprising: first, second and third inputs for receiving R, G and B signals respectively; first combining means connected to the inputs for providing an R-G signal art a first outputthereof and a B-G signal at a second output thereof; first multiplying means connected to the first output of the first combining means for multiplying the R-G signal by a factor effectively substantially equal to 0.299; second multiplying means connected to the second output of the first combining means for multiplying the B-G signal by a factor effectively substan tially equal to 0.114;; and second combining means connected to the outputs of the first and second multiplying means and the second input to provide a luminance signal Y = G + 0.299 (R - G) + 0.114 (B - G) and also to the first and third inputs to provide signals R - V and B - V.

2. A matrix according to claim 1, provided with gain circuitry connected to convert the B -Y and R V signals into U and V signals.

3. A method oftranscoding a PAL colouctelevi- sion signal in R,G,B form to P-V, V, B-Vformrcom- prising: subtracting the G signal from each of the R and B signals to give R-G and B-G signals; multiplying the R-G signal by a factor effective-fy substantially equal to 0.299; multiplying the B-G signal by a factor effectively: substantially equal to 0.114; adding the two multiplied signals and the G signal to provide a V signal; and subtracting the Y signal from each of the R and B signals.

4. Atranscoding matrixfortranscoding PALcolour television signals in P-V, V, B-Y form to R,G,B form, comprising: first, second and third inputs for receiving P-V, V, and B-Y signals respectively; first combining means connected to the inputs for adding the R-Y and Y signals to provide an R signal and for adding the B-Y and Y signals to provide a B signal; first multiplying means connected to the first input for multiplying the R-Y signal by a factor effectively substantially equal to (0.299/0.587); second multiplying means connected to the third input for multiplying the B-Y signal by a factor effectively substantially equal to (0I44/0.587);; and second combining means connected to the outputs of the two multiplying means and the second input to subtract the sum of the multiplier outputs from the Y signal to provide a G signal.

5. A matrix according to claim 4, adapted to receive Y, U and V signals, and modified by the inclusion of respective multipliers between the first and third inputs and the first combining means, and in that the multiplication factors of the said first and second multiplying means are correspondingly adjusted.

6. Amethod of transcoding a PALcolourtelevision signal in P-V, V, B-Y form to R, G, B form, comprising: adding the Y signal to each of the R-Y and B-Y signals to provide R and B signals; multiplying the R-Y signal by a factor effectively substantially equal to (0.299/0.587); multiplying the B-Y signal by a factor effectively substantially equal to (0.114/0.587); and subtracting the sum of the multiplied signals from the Y signal to provide a G signal.