JPS60197088A - Processing circuit of color video signal - Google Patents

Abstract

PURPOSE:To obtain a digital color video signal of good quality by band-limiting two color difference signals by means of a digital low-pass filter and by making a digital low-pass filter into an FIR digital filter with zero phase. CONSTITUTION:Three primary color signals read out from frame memories 47, 48 and 49 are supplied by one to a processor 50, and a digital color component signal is produced by the arithmetic. A generated luminance signal Y, color difference signals U and V are written in data memories 51, 52A and 53A, respectively. At this phase, the signals U and V are the same high-pass signal as the luminance signal. A band control of the signals U and V stored in the memories 52A and 53A is carried out by a digital low-pass filter, which has a basic construction of an FIR digital filter and zero phase.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a color video signal processing circuit for forming Y, U, and V color component signals from three primary color signals.

"Background technology and its problems" When recording a color still image signal on a digital disk, the three primary color signals R, G, and B are converted into a color component signal Y.
. FIG. 1 shows the configuration of an example of a color video signal processing circuit for this purpose.

In Figure 1, 1.2.3 is the three primary color signals R2G, B
4°5.6 indicates the input terminals to which 0.299, 0.587, and 0.114 are respectively applied.
It shows an attenuator that multiplies the primary color signal. The respective outputs of the attenuators 4.5.6 are added by the adder circuits 7 and 8, and the output of the adder circuit 8 is (0,299R+0.587G+0.114B).
A luminance signal is extracted. This luminance signal is transmitted to the delay circuit 9
and loaf (supplied to N-container 11 via filter 10).

The output of the adder circuit 8 is supplied to a phase inversion circuit 12, and the output of this phase inversion circuit 12 is supplied to adder circuits 13 and 14. The adder circuit 13 receives a red signal R from the input terminal 1.
is supplied to the adder circuit 14, and a blue signal B is supplied from the input terminal 3 to the adder circuit 14.

Therefore, the (RY) signal is taken out from the adder circuit 13, and the (B-Y) signal is taken out from the adder circuit 14. Attenuator 1 where the (RY) signal is multiplied by a coefficient of 0.877
5 and a low-pass filter 16 to the A/D converter 1γ. The (B-Y) signal is supplied to the A/D converter 20 via an attenuator 18 that multiplies it by a coefficient of 0.493 and a low-pass filter 19.

Compared to the sampling frequency of the A/D converter 11,
The sampling frequency of A/D converters 17 and 20 is 1
/3, and compared to the luminance signal Y, the number of data of the color difference signals U and V is 1/3. The digital component signals digitized by each of these A/D converters 11, 17, and 20 are stored in data memory l) 21.
, 22, 23. Data memo 1) 21,2
A color still image is formed using the data stored in 2.23, and the data stored in data memo 1, 121, 22.23 is converted to a recording signal in a predetermined format and recorded on a disk, etc. be done.

In the above-mentioned color video signal processing circuit, the low-pass filter 10 limits the band of the luminance signal to 4.8 MHz, and the low-pass filter 16.19 controls the band of the color difference signal to 1.6 JVI Hz.

Figure 2 shows the frequency characteristics 3 of the low-pass filter 16.19.
1 and a group delay characteristic 32, and FIG. 3 shows a frequency characteristic 33 and a group delay characteristic 34 of the low-pass filter 10.

As shown in FIGS. 2 and 3, the group delay of the low-pass filter 10 is 860 n sec, while the group delay of the low-pass filter 16.19 is approximately 23 n sec.
30 n sec, and the color difference signal is significantly delayed compared to the luminance signal.

A delay circuit 9 is provided to compensate for this delay.

However, as is clear from the group delay characteristic 34 in FIG.
A time difference occurs, which causes color shift. FIG. 4A is used to explain the sampling of a color difference signal, for example, a color difference signal V, where 35 shown by a broken line is the waveform of the original signal of the color difference signal, and 36 shown by a solid line is the waveform after passing through a low-pass filter. This FIG. 4A is an ideal case in which the low-pass filter has a delay of O.

As mentioned above, a group delay occurs in the analog low-pass filter 19, and therefore, as is clear from the signal waveform 3γ indicated by the dashed line in FIG. 4B, a delay occurs in the signal, and this delayed signal waveform 37 Sample data V1' is the result of sampling.

V, , V≦ is data V1.V1.V≦ obtained by ideal sampling shown in FIG. 4A. V2. If the value is different from V3, it will cause deterioration of the reproduced image quality.

Furthermore, even if a method is used in which the color difference signals U and V are sampled at the same sampling period and the same timing as the luminance signal Y, and this sampling pattern is adopted only once every three times, the sample shown in FIG. As is clear from the data, there was a problem in that the sample data V2.V3 had a different value when ideal sampling was performed, and the color tone changed. .

``Object of the Invention'' Therefore, an object of the present invention is to provide a color video signal processing circuit that can obtain a high quality digital color video signal without causing a phase delay in a low-pass filter for narrowing the band of a color difference signal. It is about providing.

"Summary of the Invention" This invention limits the band of each of two color difference signals using a digital low-pass filter, and applies predetermined coefficients to past sampling data and previous sampling data, respectively, as the digital low-pass filter. This includes a zero-phase FIR digital filter that multiplies and adds.

"Embodiment" An embodiment of the present invention will be described below with reference to the drawings. In FIG. 5, which shows the configuration of an embodiment of the present invention, three primary color signals R, G, and B of a color still image signal are supplied to input terminals indicated by 1, 2, and 3, respectively. These three primary color signals pass through low-pass filters 41, 42, and 43 respectively.
It is supplied to D converters 44, 45, 46. The low-pass filters 41, 42, and 43 have a pass band of, for example, 4.8.
MHz, A/D converter 44,454
6, digitization is performed using a sampling frequency such as 3 f sc (fsc: color subcarrier frequency) r 41 sc. A/D converter 4
The digital three primary color signals from 4.45.46 are stored in frame memories 4.gamma. and 48.49, respectively.

The digital three primary color signals stored in the frame memos IJ 47 , 48 , 49 are read out to the processor 50 . The Zorocessor 50 converts the digital three primary color signals into y,
Processing for converting u and v into digital color component signals, processing for limiting the bands of the digital color difference signals U and V, and processing for writing the band-limited digital color difference signals U and V into the data memory are performed.

One sample of the three primary color signals read out from each of the frame memories 47, 48, and 49 is supplied to the processor 50, where a digital color component signal is formed by the following calculation.

Y=-0,299R+0.587G+0.114BV=
0.877 (RY) U=0.493 (B-Y) The luminance signal Y thus formed is written into the data memory 51, the color difference signal U is written into the data memory 52A, and the color difference signal (2) is written into the data memory 53A. At this stage, the color difference signals U and V are wideband signals identical to the luminance signal. -Next, data memo IJ 52 A
Band-limiting of the color difference signals U and 2 stored in 53A and 53A is performed by a digital low-pass filter. This digital low-pass filter has the basic structure of an FIR digital filter and has zero phase. This digital low-pass filter with zero phase will be explained below.

As an example, for a 9th order FIR digital filter with linear phase, hμ=h9-1', (μ=0, 1, =
-, 9). The frequency transfer characteristic of the digital filter is given by substituting ZK ejoT into H(z). (ej"T - ejΩ), the frequency transfer characteristic of the above-mentioned 9th-order digital filter is as follows. Also, regarding the 9th-order FIR digital filter using future data, the frequency transfer characteristic is expressed as When these two FIR digital filters are connected in cascade, the frequency transfer characteristic is as follows.As is clear from this frequency transfer characteristic G (Ω), there is no phase term, so the digital filter has zero phase. can be configured.

Since we are assuming a digital low-pass filter with maximum gain in direct current and zero phase, we will perform normalization again. The above frequency transfer characteristic G The amplitude characteristic of the frequency transfer function of the digital low-pass filter to be obtained is as follows.The transfer function of the digital low-pass filter to be obtained is, however, hμ-ri, (μ=0, 1,...) It is.

When the degree is 9, each coefficient of H(z) H(zl) is i ≦9 ak= 1! i-, h, +kh.

The transfer function of the digital low-pass filter is expressed as follows.

1 1 As shown in Figure 6, the above formula connects 18 unit delay elements in cascade to the input terminal 61, and draws the output from between the stages between the 9th unit delay element and the 10th unit delay element. A Mari coefficient multiplier is connected to the tapped taps, and the taps pulled out from the front and rear stages are symmetrically connected to a, a, a2, .・・・・・・＋
``Nine coefficient units are connected and the output of each coefficient unit is added to the adder circuit 62.
This corresponds to the configuration of a digital filter in which the added output is derived as an output terminal 64 via a coefficient unit 63 with a coefficient bo. The purpose of the coefficient multiplier 63 is to multiply the DC gain value of the denominator term.

Generally, the above equation also holds when the order is N (N is an odd number), and the amplitude characteristic at that time is as follows.

The processor 50 in one embodiment of the present invention performs the filter operation of the above-mentioned digital row filter as shown in FIG. This filter calculation is performed on each of the color difference signal U stored in the data memory 1) 52A and the color difference signal ■ stored in the data memory 52B, and each of the filter-calculated color difference signals U and ■ is It is made into an arterate and written to memory 52B and 53B.

In addition, in the filter calculation, N sample data are required before and after the current sample data, so N data are added before and after the first still image data, and the still image data N pieces of data are added after the last one. As this added data, in addition to the original data located at the outer periphery of the color still image, an intermediate value of the quantization level of the sample data or an average value of the screen may be used.

A luminance signal stored in data memory 51.

The color difference signals U and V stored in the data memos IJ 52B and 53B, respectively, become digital color video signals corresponding to one color still image.
Color video signals read from these data memories are made into recording signals in a predetermined format and recorded on a recording medium such as a digital disk.

In the embodiment of the present invention described above, the processor 50 forms the Y, U, and V signals and limits the band of the color difference signal. However, from the RIG and B signals, Y, U,
It is also possible to adopt a configuration in which the V signal is formed by a digital matrix circuit, and then the band of the color difference signal is limited by a digital low filter having a lead wired configuration as shown in FIG.

[Effects of the Invention] According to the present invention, it is possible to solve the problem of the phase shift between the luminance signal and the color difference signal, especially on the high frequency side, which occurs when band limitation is performed using an analog low-pass filter. The digital low-pass filter according to the present invention has a zero-phase characteristic, and a delay circuit for phase adjustment can be omitted in the signal path of the luminance signal.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an example of a conventional color video signal processing circuit, Figs. 2 and 3 are graphs used to explain a conventional analog low-pass filter, and Fig. 4 is a block diagram of an example of a conventional color video signal processing circuit. FIG. 5 is a block diagram of one embodiment of the present invention, and FIG. 6 is a block diagram showing the basic configuration of a digital low-pass filter in one embodiment of the present invention. 1.2.3 --- Three primary color signal input terminal, 44
゜45.46 ・・・・・・A/D converter, 50
...Zorosetuza, 51.52A, 52B,
53A. 53B ・・Data memory. Agent Masato Sugiura

Claims (1)

[Claims] A digitized luminance signal and two digitized color difference signals are supplied, and each of the two color difference signals is band-limited by a digital low-pass filter. 1. A color video signal processing circuit characterized in that it is a zero-phase FIR digital filter that multiplies previous sampling data by predetermined coefficients and adds the results.