JPS6062292A - Y/c separating circuit - Google Patents

Abstract

PURPOSE:To obtain an adaptive type two-dimensional Y/C separating circuit and perform Y/C separation with high precision by performing linear Y/C separation in plural directions on a two-dimensional array, and using a weighting coefficient according to the precision of the separation to obtain a weighted mean. CONSTITUTION:A digital color composite TV signal is separated by an Y/C separating circuit into a luminance and a chrominance signal. This separating circuit decides on plural directions based as centers upon sample data to be separated on the two-dimensional array consisting of plural sample data. Then, the Y/C separation is performed with other sample data positioned in those plural directions and sample data to be processed, and the luminance or chrominance signal obtained by the separation is used to obtain evaluated values in the respective directions. Then, weighting coefficients in the plural directions are obtained on the basis of the evaluated values. Those weighting coefficients are used to separate and output the weighted mean value of Y/C separation outputs in the plural directios, thus performing the Y/C separation with high precision.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention This invention relates to a Y4 separation circuit that separates a luminance signal and a carrier color signal from a composite color television signal converted into a digital signal.

"Background technology and its problems" Conventional Tigitalun. As a separation circuit, the color subcarrier frequency fso (3.58M1 (z
), a bandpass filter with a passband centered at
Vertical direction composed of comb-shaped filters. Separation circuits, or two-dimensional six-separation circuits in which both are connected in series are known.

However, it is one-dimensional. Separate circuit and 2D. Established based on the statistical characteristics of input and output data, regardless of the minute circuit) 1
Conventional fixed-coefficient sand separation filters are effective for data with average characteristics, but have the drawback of large errors for data with characteristics. In other words, the coefficients are determined by the data of stationary characteristics, which account for the majority of the coefficients.
Where there are small patterns or changes, the accuracy of Vo separation is extremely reduced.

In order to solve this problem, the correlation between the lines is detected from the absolute values of the color videotakes of the three lines. An adaptive Y//c/1 circuit has been considered that performs vertical Vc separation using a simple average value of . However, since the correlation is detected only at the top and bottom in the vertical direction, and the average value is used as the separated output, there is a drawback that the accuracy/ba of VC separation is low.

``Object of the Invention 1 The object of the present invention is to provide an adaptive two-dimensional separation circuit capable of performing high-precision Y/C separation.

1 of the Invention (Existing Summary 1) This invention defines a plurality of directions centered on a sample take in which separation processing is performed in a two-dimensional array consisting of a plurality of sample data, and Perform Y/c separation processing using the sample data and the sample data to be processed, and use the luminance signal or color signal obtained by this separation processing to obtain evaluation values for each of a plurality of directions.This evaluation value The weight gain for each of the plurality of directions is obtained based on the coefficient f:, and this weight gain is calculated for the plurality of directions using each family of coefficients.The weighted average value of the separated outputs is obtained as the separated output.

"One Embodiment" Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In one embodiment of the present invention, an NTSC composite color television signal is converted into a 1-sample, for example, 8-bit digital color television signal using a sampling pulse of frequency 4f having a predetermined phase. Fig. 1 is an enlarged view of a two-dimensional array of sample takes in one embodiment of the invention, where the horizontal line is the line L - within one field.
2, L −+ , , Lo, Ll, and I−2 are respectively shown, and the vertical line indicates the horizontal sampling position S −
2+ S -1+ So+ SI+ s2i are shown respectively.

The sampling position interval is the time TS(-”4 J's
. ).

When the sampling frequency is 4fsc, U・rnI, 0
If the sample data (white square) r: M (not shown) at the sampling position SO is the data to be separated by Y//c, then 1 for this sample data e
The phase of the sub-gear in the general feed color signal is as shown in FIG. The white circular dots located 4 TsL- apart on the same line and the same dots /L (+
＞White device installed every other inch] Friendly circle 1
If the throat take is in the same phase as the sample take e2.
), black circular dots located every 2T'S at the same sample position, black circular dots located every other line at the same sample position, and black circular dots located every other line from this adjacent line with opposite phases. In addition, the white knee-shaped dots have a phase shift of ■, and the black triangular dots 1- have a phase shift of -T.

In this example, sample take e has a distance f(31
＜＞Sample data a where the subcarrier is in opposite phase nearby,■
), c, and d as shown in the following formula:
) -t e) / 2 ・ ・ ・ ■Y3-(c
+e) / 2 ■ Y, = (d 1 e ) / 2 ・ ■These Y/
/c separation is a one-dimensional process, and in order to evaluate the accuracy of husband's Y/10 separation, in-phase sample data a' + b' + c whose distance is closest to sample take e is
It is expressed by the following evaluation function using '+d'.

P-] / l a'-el ■ P-1/ l b'-e l -■ p=1/lc'-el ・・・■ P=1/ld e l ■ Equations from ■ to ■ indicates the degree of correlation and stationarity between the sample take e and each of the surrounding sample takes a, l), c, and d. Size,■
In formulas 2 to 2, if the denominator term is 0, the evaluation function is set to a sufficiently large number. To obtain the evaluation function, you may square the difference instead of taking the disjoint value.

The evaluation functions P, , P2 . From the P3° bow, Q weighting is calculated using the following formula: w2°W3
, W4 is seen.

w, = p, 7 (P, -1-P21- P, 1-1
- P, ) -0w2= p, , / (P, I-P
2+P3+ P, ) -■W:i -P:+/ (+
)+ 4- P2-1- P3-1- P4) ,,
, ,,, QWs-P4/ (PI t-P2-1-
P3-k P4 ) -■Here, (w, , w, ,
, w3. w,, (+) and (w, −l−
w2+w3-l-W4=1). Using this weighted 1'' coefficient, the luminance signal of sample data e is obtained as a weighted average of sample data a, b, c, and d.

In other words, Y=W+-χ1+w2-Y2+ w3-Y3+W<”Y
＜ −■. Furthermore, the color signal is separated by subtracting the luminance signal determined in this way from the sample data e.

The following processing is performed for each sample data.

Here, the color signal may be first separated by the same processing as described above, and then the luminance signal may be separated.

FIG. 2 shows the configuration of an embodiment of the present invention that performs the above-described processing, and an input terminal 1 is supplied with a digital composite NTSC color television signal sampled at a frequency of 4 fsc.

This input signal is transmitted through delay circuits 2゜3.4. each having one horizontal period.
5 in series and a delay circuit of J sample period (T8) each; 6. γ, 8, and 9 are connected in series. Delay circuit 2
delay circuits 10, 11, 1 each having an output of one sample period;
2.13 in series connection. Delay circuits 14, 15, 16, 17 whose outputs from the delay circuit 3 each have one sample period
, 18° 19.20 in series connection. Delay circuits 22, 23. each output of the delay circuit 4 has a one sample period.
24.25 in series connection. Delay circuits 26, 27, 2.8 in which the output of delay circuit 5 is 1 sample j, respectively.
．． 29 series connections.

When the output of the delay circuit 17 is sample data e, the output of the delay circuit 3 is the sample data C' of lines L and O, and the outputs of the delay circuits 15, 19, and 21 are sample data c, a, a'. is taken out.

Further, the delay circuit 9 outputs the line L2 pull data d', and the delay circuit 13 outputs the line L2 pull data d'.
sample data d is taken out, and data b of line I, -1 is taken out to the output of the delay circuit 25.
Sample 0 data b' of line L-2 is taken out at the output of 9.

The outputs of the delay circuits 17 and 19 are supplied to an adder circuit 31, and the output of the booster circuit 31 is passed through a multiplier 35 of 7 to form a luminance signal Y1 expressed by the above equation 0. The outputs of the expansion circuits 17 and 25 are supplied to the adder circuit 32, and the output of the adder circuit 32 is 7 to the power fi? : A luminance signal Y2 expressed by the above equation 0 is formed by passing the luminance signal Y2 through the device 36. Similarly, the adder circuit 33 and the multiplier 3γ form the brightness multiplier Y expressed by the 0 equation, and the adder circuit 34 and the multiplier 38 form the luminance multiplied Y expressed by the 0 equation. These multipliers 35.36.37.
38, it is realized by a configuration in which 1 hinotonft is stored in the futon register.

The subtraction circuit 41 forms the output of (a'-e), and the subtraction circuit 42 forms the output of t(b'-e) -jl
4I: 11'41. t；'iノ・1. ,9+D鴇゛IiTl1mAIJI/' , ]'h, (r' −
ρ) is formed, and the subtraction circuit 44 produces an output of (cl'
-C) outputs are formed. These subtraction circuits 41°4
ROM45.46゜47.48 that generates the reciprocal of the absolute value of these outputs for each output of 2.43.44
is provided. Therefore, ROM 45.

As each readout output of 46, 47, and 48, the above ■
Evaluation function P expressed by formula, ■formula, ■formula, and ■formula, ,
I)2. I)3. P, is obtained.

Addition circuits 5152.53 calculate the sum of evaluation functions (p
, -1-p2+ P3-1- P4) are calculated, and this sum is supplied to dividing circuits 61, 62, 63, and 64 as denominator inputs. As the input of the division circuit 610 numerator,
The evaluation function 21 is supplied from the ROM 45, and the weighting coefficient W1 expressed by the above-mentioned equation 0 is calculated.

As the input of each numerator of the division circuits 62, 63, and 64,
Evaluation function P2 from ROIv'l 46 , 47 , 48
．． P...

1 is supplied, and the weight + one-digit coefficient w2. w3. w is calculated.

For weighting, coefficient WI and luminance signal Y are supplied to the multiplier 71, and for weighting, coefficient w2 and luminance signal Y2 are multiplied by z:;
γ2 (supplied, the following coefficient w3 and the luminance signal Y,
is supplied to a multiplier 73, and a weighting coefficient W4 and a luminance signal Y4 are supplied to a multiplier 74. These multipliers 71
, 72, 73. The outputs of γ4 are added by adders 75, 76, and 77, and the luminance signal Y expressed by the above equation 0 is separated at the output terminal γ8. Finally, in the subtraction circuit 81, the luminance signal Y, which has been subdivided by 11, is subtracted from the sample take e, and the carrier color signal C is taken out at the output terminal 82.

Unlike the above-mentioned embodiment, subcarriers in the upper and lower four directions are
It is also possible to use a plurality of other sample takes of opposite phases with i+' in the direction of the sample to perform all of the processing for one dimension 'f//c', the calculation of the evaluation function, and the calculation of the weighted average. .

1. This invention is 4f. 71 can also be applied when digitized at any sampling frequency other than the above. In this case, the vertical Y/C separation process uses a sampling frequency of 4 f.

However, the Y/c separation process in the horizontal direction is performed by the configuration shown in FIG. FIG. 3 shows the configuration for obtaining the luminance signal Y1. In other words, when the sampling frequency is 4f, the processing corresponding to Yyc fraction 1 shown by the equation C0 is performed.

Y, = (a + 2 e + c) / 4 --
(The current sample data targeted for f3VC separation can be expressed as e = Y, +C, = Y, 10 K sin θ. The data q is expressed as f = Y, −l−K sin (θ−Δθ)・@kuY, −1−K sin (θ−Δθ)・@.Here, Δθ is one sample interval It is the phase difference corresponding to Ts. From the above equation, j"2cosΔ0・e+') = Yl-1−K sinθ cosΔθ−f(cos
θsin Δθ −2cos Δθ fist Y+−2K sin
θ'cosΔθ+Yl−1−KsinθcosΔθ10K
cosθsinΔθ=(22cosΔθ) Yl --
-@ is i:I. Therefore, Y, = (j'-2cosΔθ”e+))/(2-2c
osΔθ) °0. Corresponding to this equation 0, as shown in FIG.
) The luminance signal Y+ can be separated by the poMs 93 and 96 that generate f, the subtraction circuit 94, and the addition circuit 95.

1 Effect of the Invention According to the invention, the conventional one-dimensional. Separation circuit or heavy I' (VC separation filter in the -+ direction and Y/
Separated type secondary with C separation filters connected in series>if Y
It is possible to realize a river-separation type two-dimensional six-separation circuit with a higher degree of separation than the /Ce separation circuit.

This invention performs one-dimensional VC separation in multiple directions on a two-dimensional array, and uses a coefficient to obtain a weighted average according to the separation accuracy, so three samples in the vertical direction are The separation accuracy can be improved compared to a conventional adaptive Y/C separation circuit that uses a simple average value of two sample takes that are determined to have a strong correlation among the takes.

[Brief explanation of the drawing]

Fig. 1 is a route map showing a two-dimensional arrangement of sample takes used to explain one embodiment of the present invention, Fig. 2 is a block diagram of one embodiment of the invention, and Fig. 3 is another embodiment of the invention. FIG. 2 is a block diagram used for explanation. 1. Input terminal for digital composite color television signal, 2, 3, 4, 5--1 line delay circuit, 78. Output terminal for luminance signal, output terminal for 82 color signals. Agent Masato Sugiura

Claims (1)

[Claims] Sample data that is subjected to separation processing on a two-dimensional array consisting of a plurality of sample data in a separation circuit that separates a luminance signal and a carrier color signal from a digital composite color television signal. A plurality of directions are determined with a center at An evaluation value for each of the plurality of directions is obtained using the color signal, and based on this evaluation value, a coefficient is obtained for each of the plurality of directions, and this weight is calculated using each of the coefficients. Directions in multiple directions above. The weighted average value of the separated outputs is obtained as the separated output. Separation circuit.