JPS58129890A - Adaptive type luminance signal and chroma signal separation filter - Google Patents

Abstract

PURPOSE:To separate luminance signal from chroma signal with very clearly in characteristics,by providing a circuit by newly selecting an input signal where the absolute value of the sum of the signal with a signal selected by a half period earlier of a sub-carrier is minimized. CONSTITUTION:A composite signal 1 is converted into a digital video signal 3 at an A/D converter 2. The signal 3 is inputted in separation filters 4A-4D of lateral, longitudinal and oblique correlation via a delay circuit 19. Chroma signals 5A-5D separated by operating the correlation of each direction are inputted to a discrimination circuit 6. The chroma signal 9 selected by a half period earlier of the sub-carrier is inputted to the circuit 6. The circuit 6 outputs a signal minimizing the sum of the signal with a signal 9 among the signals 5A-5D. In this selection algorithm, the chromaticity (further, luminance) is separated accurately to a pattern where the luminance is rapidly changed with a gradual change of chromaticity.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention separates a luminance signal and a chrominance signal from a composite power-television signal with excellent characteristics.

This invention relates to an adaptive luminance signal/chromaticity signal separation filter.

[Technical background of the invention and its problems]

In the NTS C system, to transmit the color tL signal, it is modulated with a subcarrier, and the 2.
It is superimposed on the 1 MHz to 4.2 MHz portion. There is a relationship between the subcarrier frequency Fsc and the horizontal frequency FH, and the 1Iif signal and the chromaticity signal are frequency interleaved with each other. The phase of the tjls+carrier wave is located one line apart and has an opposite phase (conventional luminance signal and chromaticity signal separation filters use this relationship to separate b9).

In other words, the NTSC signal is divided into one horizontal period TH (-1/PH)
By adding this signal to the signal before the delay, the m* transmission (chromaticity signal) is canceled and only the luminance signal is obtained, and by taking the difference between the two signals, the luminance signal is canceled and the chromaticity signal is obtained. f1! Only the issue number is obtained. However, this method requires that the signals are equal 1-)in apart, i.e. 22
This is based on the assumption that the same brightness and chromaticity exist throughout the interior, but this generally does not hold true. In other words, only patterns with small vertical changes (strong correlation) can be accurately separated by this filter.

Further, contrary to the above-mentioned filter, there is a method described below as an example of a filter that separates luminance and chromaticity using correlation in the horizontal direction of the screen. By delaying the N'I'8C signal by half the subcarrier period Tsc/2 (= 1/2Fsc) and adding this to the signal before the delay, the carrier wave (chromaticity signal) is canceled. and only the luminance signal is obtained.
Also, this is a method of obtaining only the chromaticity signal by taking the difference. In this case, the same bright color and
It is assumed that chromaticity exists. Considering the band of the chromaticity signal here, the ■signal is 1.4MHz, and the Q signal is 0.
．． Since the frequency is 5 MHz, the half period of the subcarrier (7, 1
chromaticity is considered to be almost constant, but since the luminance signal band is as wide as 4.2 MHz, # degree cannot be said to be constant even within this time range, and the above assumption This does not hold true for Hiro in general. This means that, contrary to the conventional example, separation by this filter is properly performed only for patterns that have a small change in the horizontal direction (strong correlation).

In addition, for patterns that vary greatly both in the vertical direction and in the horizontal direction, such as a nine-striped pattern running diagonally, it is necessary to separate luminance and chromaticity using the correlation in the diagonal direction.

Furthermore, each of the chromaticity signals separated using the above-described horizontal, vertical, and diagonal correlations is multiplied by a weighting coefficient to obtain the output color tL1! There is also a method of using a multi-directional correlation, but this method has a certain limit to its performance because the advantages and disadvantages of each separation filter are superimposed at the same time. As a result, it also has the disadvantage that the resolution of the entire screen is significantly reduced.

Luminance 1-1 with constant frequency characteristics as described above
．． With a chromaticity signal separation filter, it is impossible to accurately separate patterns that change in various ways. On the other hand, in the field of image processing using computers, an adaptive method is used to determine the brightness and chromaticity of a single point on the screen by detecting surrounding patterns and using an optimal separation filter. . According to this method, luminance and chromaticity can be separated accurately regardless of the pattern.

However, it is difficult to apply this method to consumer television receivers, which require real-time processing of signals and are subject to large cost constraints, in terms of signal processing speed and node square footage.

[Purpose of the invention]

The present invention has been made with this in mind, and its purpose is to provide adaptive luminance signal/chromaticity signal processing using relatively simple algorithms and hardware! 7. It is about providing a unique service.

[Summary of the invention]

The present invention uses the O correlation in each of the horizontal, vertical, and diagonal directions to separate the signals, outputs them by -1- in front of the nine chromaticity signals, and selects the one whose Fsc nine chromaticity signal and O sum are closest to zero. This provides a Jil11 filter for luminance and chromaticity signals.

[Embodiments of the invention]

The present invention will be explained in detail below. Part 1II is a schematic diagram showing one embodiment of the present invention. N'lC composite signal 1
is sampled and digitized by the A/D converter 2 and converted into a digital video signal 3. The sampling period Ts is 1/4 of the subcarrier period 'rsc (
Ts = 1/4Fsc). A reference clock 17 having a period of 1 and 1 is generated by the clock generator 16
It is input to each circuit as a reference for operation timing. Here, if the NT8C@ combined signal l is expressed as V(t), the digital video signal 3 is the value of T8m/pull of V(t), so excluding the finite word length error due to digitization, V
(nTs) (rl=1.2.-)
Obushi2i! This figure shows the correspondence between the scan @2-1, the carrier wave 2-2, and the sample point 2-3 on the television screen, with l being centered around the sum/glue point of n=nQ. Sample period Ts
The relationship between TH and the horizontal period TH is TH=910TS, and the points directly above and below the n6th sample point on the screen are n6-910. This is a sample point of ng+910 writing times. Digital video signal 3 (V(nT)
s) ) has a unit delay time of Ti and a total delay time of 18
26Ts (-2Ta+6Ts) is input to the delay circuit 19. Here, the plurality of digital video signals 920, 21, 22, and 23, which have undergone a predetermined delay, are input to horizontal, vertical, and diagonal correlation separation filters 4A to 4D, respectively.

Here, the digital video signal 20 input to the four horizontal correlation separation filters is V(nQTs), V[(nQ −
2) Ts), V((n・+2)TI), digital video signal 21 input to the vertical correlation separation filter 4B
is V(n, Ts)V((no 910)TJ, V
t(no+910)Ts), diagonal correlation $ filter 4CK input digital video (1't22 HAV
(nQT)V((n(,-907)TJ,V(
(no+907) 6 times at TJ, diagonal correlation @ digital video signal 23 input to filter 4D is V(n0Ts
), V((no -913)TJ, V((no
+913)Ts). Note that n6=913. n
In order to separate the chromaticity signal at the 0th sample point, the first
Separation filters 4A for horizontal, vertical, and diagonal correlation in the figure
~4D performs the following calculation.

Horizontal correlation chromaticity signal 5 which is the output of the horizontal correlation separation filter 4A
A (represented by C(A)) is obtained by the calculation of T...(1). Sample points n0-2 to n, +2
If the same brightness and chromaticity exist in the range of , the chromaticity signals at sample point no-2 and n0+2 have opposite phases with respect to sample point n0, so according to equation (1), the synangle points! The correct chromaticity at 1o is separated.

Vertical correlation chromaticity signal 5 which is the output of the vertical correlation separation filter 4B
B(C(87) is obtained by +V((910)Ts7 ko・(2). Sample point n・−9IQ, II,.

If the same brightness and chromaticity exist at the three points n@, +910, then n6-910 and n(, +
In step 910, the chromaticity signal is in reverse phase, and the correct chromaticity signal at the synangle point n0 is separated according to equation (2).

Diagonal correlation chromaticity signal 5C (represented by C (C)) which is the output of diagonal correlation separation filter 4C
xoTs) + [V((me 907)T
s)+V 5L(no+907)TS) ] −(3)
) can be obtained by the operation of Here, the sample pull point n@-907+ no v no +907 was selected as the calculation point.

This is because the angle formed by the line connecting these three points and the scanning line 2-1 is relatively close to 45°. Sample point n.

If the same brightness and chromaticity exist at n, -907, Il, +907, the chromaticity signals at sample point n, -907 and n, +907 will have opposite phases. According to equation (3) K, the synangle point n
A correct chromaticity signal at 0 is obtained.

The diagonal correlation chromaticity signal 5D (represented by C(2)), which is the output of the diagonal correlation separation filter 4D, is +V[(no+1
i13) It can be obtained #) by the calculation of TJ J =A4).

This is different from the diagonal correlation separation filter 4C in that it is separated by the correlation direction, and sample point n is the same. , n, ) -913,
If n6=913 and the same luminance and chromaticity exist, a correct chromaticity signal at the t-angle point n6 can be obtained according to equation (4).

The above is an explanation of the operation of each separation filter 4A to 4D. Note that since all of these filters have direct phase characteristics, they do not affect the phase of the chromaticity signal showing two hues.

The chromaticity signals 5A to 5D separated by correlation calculation in the horizontal, vertical, and diagonal direction are input to a determination circuit 6. Accordingly, the chromaticity signal 9 selected 312t sample time (2Ts) earlier is also simultaneously input to the determination circuit 6. The chromaticity signal 9 is the chromaticity signal 7 selected by the judgment circuit 6,
The signal can be passed through a delay circuit 8 having a delay time of T3. The judgment circuit 60 function outputs the one whose chromaticity signal 81 is closest to zero among the chromaticity signals 58 to 5D separated by the correlation calculation in each direction and which is selected before the chromaticity signal 9. It is to be. This selection algorithm utilizes the narrowband nature of the chromaticity signal, and is a feature of the present invention.
It is explained as follows. Since 2Ts is equal to the half period of the subcarrier O, if the chromaticity is constant in this 0 time range, the O chromaticity signal at this moment and the chromaticity signal 2 sample times (2Ts) ago are in reverse phase. In the relationship, Nada, its Wa Mori zero.

Chromaticity signal (I signal - 15M) h, Q signal α5M
Hz) and 2Ts (7,16MHxk phase), it is considered that the chromaticity within this time is almost constant.

In other words, the current chromaticity signal is 2Th(-sc)
According to the algorithm, the sum of the previous chromaticity signal is close to zero. The separation is performed accurately (kana). The chromaticity signal 7 selected by the judgment circuit 6 in this manner is inputted again to the judgment circuit 6 via the above-mentioned 90-degree circuit 8, and the above-described judgment operation is repeated.

The above is an explanation of the adaptive operation.

The chromaticity signal 9 is output from the 0-fi D/A converter 1
In addition to being outputted as an analog color flI number 1m through 1, it is also input to a subtraction circuit 10. A digital video signal 24 obtained by changing the phase of the digital video signal 3 to the chromaticity signal 9 by a delay circuit 15 is inputted to the subtraction circuit 1o. By this subtraction, the luminance signal 18 is separated and outputted as an analog luminance signal 14 via the D/A converter 12. In the determination circuit 6, it is also necessary to provide an output priority order in the case where a plurality of selections satisfy the selection conditions. Conversely, if all of the sums of the nine chromaticity signals 5A to 5D and the chromaticity signal 9 input to the 1 judgment circuit 2 circuit 6 are more than a certain set value from zero, a predetermined chromaticity signal determined separately is determined. It is necessary to take measures such as setting the chromaticity signal separated by the separation filter as the output of the determination circuit 6, or setting the output of the determination circuit 6 to zero (determining that there is no color). The above is a detailed explanation of the present invention shown in Figure jI1.

[Departure F! Effect of A]

As mentioned above, the feature of the present invention is that the sum of the nine chromaticity signals, which is separated using the correlation in the horizontal, vertical, and diagonal directions, and the chromaticity signal outputted just -1- before has the highest FSC. The goal is to select something close to zero. This method utilizes the narrowband nature of the chromaticity signal, and its nine strengths do not require the use of complex pattern recognition algorithms, but instead use relatively simple hardware. There are K things that the separation filter can accomplish.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, and FIG. 2 is a diagram showing the relationship (C) between the scanning line, the subcarrier phase, and the sample point. 1...NT8C composite signal, 2...A /1) Konno (
- data, 3...digital video signal, 4...horizontal correlation separation filter, 4B...vertical correlation separation filter, 4c, 4I)...diagonal correlation separation filter, 5 people...
- Horizontal correlated chromaticity signal, 5B... Vertical correlated chromaticity signal. 5C, 5D... Diagonal correlated color lL slight signal 6... Judgment circuit, 7... Chromaticity signal, 8... Slow circuit, 9...
Color if multiplied signal ko10...subtractor, 11.12
...D/A controller (-ta), 13... Analog chromaticity signal, 14... Analog luminance signal, 15... Delay circuit, 1
6... Black and white generator, 17... Reference clock, 1
8,... Luminance signal, 19... Delay circuits 120-24.
...Digital video signal, 2-1...Scanning line, 2-2
...Subcarrier, 2-3...Sample point.

Claims (3)

[Claims] (1) A sampling circuit that inputs the color television signal and samples the signal at a predetermined sampling period Ts synchronized with a subcarrier; A delay circuit that delays a total of more than twice one horizontal period TH, where Ts is a unit delay time, and a plurality of nine color television badges that are each delayed by a predetermined time by a different time through this delay circuit. A plurality of chromaticity signal separation filters each having different frequency characteristics, which inputs a signal and separates a chromaticity signal by multiplying each wrinkle signal by a predetermined weighting coefficient and adding the same; This is a circuit that receives nine or more chromaticity signals separated by a filter and selects one of the signals, and the absolute value of the sum of the signal selected half a period before the subcarrier is the circuit that selects one of the signals. A selection circuit that selects the minimum input signal, and a chromaticity signal selected by this selection circuit is subtracted from the output signal of the delay circuit having the same delay time as the selected signal, thereby determining the luminance. Adaptive luminance signal/chromaticity signal component 111 characterized by comprising a subtracter to obtain a signal! filter. (2) One of the multiple chromaticity signal separation filters is the chromaticity signal C
It uses correlation in the horizontal direction of the screen to separate I, and is a calculation to obtain the color ff1 (No. 1 q), *
One of them uses the correlation in the vertical direction of the screen to separate the chromaticity signal C snow, and the calculation required to obtain the chromaticity signal C2 is large. Ei's adaptive part Luminance signal/chromaticity (No. 1 separation filter. (3) Multiple 0 colors J [signal separation filter 01 uses the horizontal s function of iii* for rounding to separate the chromaticity signal C1, and the calculation to obtain the chromaticity signal C1 is,
The other one uses the vertical correlation of the rounded Klli plane to separate the chromaticity signal C8, and there is an operation to obtain the chromaticity signal C1, and the R01 separates the color **+C. in order to ii
The 90 calculation that uses the correlation in the diagonal direction (upper right to lower left) to obtain the chromaticity signal C8 is +V1n, Ts-
k(TH-m)) 2mTs (T)I for j, 4
01 uses correlation in the diagonal direction of the screen (lower right to upper left) to separate the chromaticity signal C4.
The rounding operation to obtain +Vln, Ts-k(Tx+m))
The adaptive luminance signal/chromaticity signal separation filter according to claim 1, characterized in that: