JP2001145120A - Video signal separating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a video signal separating device, which is free of deterioration in picture quality by discriminating whether a video signal is of the NTSC standard and adopting a linear (horizontal) separation system, when it not a video signal separating device, which separates a video signal into a luminance signal and a color signal modulated with a subcarrier. SOLUTION: A linear YC separating circuit 31 separates a video signal supplied from a video input into a luminance (Y) signal and a color (C) signal through a band-pass filter BPF, a two-dimensional YC separating circuit 34 separates the video signal supplied from the video input into a Y signal and a C signal via a subtracter, and an NTSC standard discrimination unit 23 adopts a YC separation system corresponding to horizontal or vertical correlation via an AND gate 24, when video signal is of the NTSC standard and a linear YC separation system via the AND gate 24 when not to select a proper YC separation system even for a video signal, whose synchronism is disordered owing to variation in tape speed, thereby preventing the image quality from becoming worse.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal separating apparatus for separating a luminance signal contained in a video signal and a chrominance signal modulated by a subcarrier.

[0002]

2. Description of the Related Art Today, color video technology such as television has been developed, and various systems for processing video signals have been adopted. Normally, the video signal is in NTSC format (National
Television System Committee standerd)
The video signal includes a luminance signal (Y signal) indicating luminance, a chrominance signal (C signal) modulated by a subcarrier (subcarrier), a synchronization signal for horizontal and vertical synchronization, and the subcarrier (sub). Carrier), and a color synchronization signal (color burst) having the same frequency as the carrier. For this reason, on the receiving side (receiver side) that receives and processes the video signal, it is necessary to separate and process the various signals.

That is, the receiving side supplied with the video signal separates the synchronization signal included in the video signal and then separates the luminance signal (Y signal) and the chrominance signal (C signal). Conventionally,
The following method is used for separating the luminance signal (Y signal) and the chrominance signal (C signal).

(A) First, a band-pass filter (BP)
F). FIG. 7 is a block diagram illustrating this separation method. The video signal from which the sync signal has been separated is
It is supplied to a band pass filter (BPF, Band Pass Filter) 1 and a subtractor 2. Bandpass filter (BP
F) 1 is a subcarrier in the video signal
(3.58 MHz), so that only the color signal (C signal) is passed, and the color signal (C signal) is output as the output of the band-pass filter (BPF) 1.
Get. On the other hand, the luminance signal (Y signal) is obtained by subtracting the chrominance signal (C signal) from the video signal by the above-described subtractor 2.

However, the above-mentioned band-pass filter (BP)
In the method using F) 1, when the luminance signal (Y signal) includes the same frequency (3.58 MHz) as the subcarrier (subcarrier), a problem of cross color occurs in the color signal (C signal). That is, since the band of the luminance signal (Y signal) included in the video signal is in the range of 0 to about 4.5 MHz, when the luminance signal (Y signal) includes the frequency of 3.58 MHz, the color signal (C signal ) Is mixed with a luminance signal (Y signal). In such a case, a problem of a cross color in which a color is formed in a portion where there is no original color occurs.

Also, for example, in the case of a vertical stripe image, if the above-mentioned frequency (3.58 MHz) component is included in the edge portion, the image quality is deteriorated. (B) On the other hand, FIG. 8 is a diagram showing a method of separating a luminance signal (Y signal) and a chrominance signal (C signal) by a comb filter (hereinafter, referred to as a YC separation method). 5 and a bandpass filter (BPF) 6. However, the band pass filter (BPF) 6 is a filter for removing noise.

The video signal from which the synchronizing signal has been separated is supplied to a line memory 3 and a subtractor 4, and the line memory 3 holds a video signal for one line. The subtracter 4 subtracts the video signal (the video signal one line before) held in the line memory 3 from the video signal (current video signal) supplied from the video input, extracts a color signal (C signal), A color signal (C signal) is output via a pass filter (BPF) 6. On the other hand, a luminance signal (Y signal) is converted from a video signal output from the line memory 3 to a color signal (C signal).
Signal) and is processed by the subtractor 5.

The extraction of the color signal (C signal) by the subtractor 4 is based on the following formula. That is, NTSC
In the standard, in order to prevent the luminance signal (Y signal) from affecting the color signal (C signal), a subcarrier (subcarrier) is used.
The relationship between the frequency fSC and the horizontal scanning frequency fH is set as follows.

Frequency fSC = 455 · fH / 2 (1) The above equation (1) shows a feature that the color signal (C signal) included in the video signal of the adjacent line is inverted by 180 °. Therefore, when the difference between the video signal one line before and the current video signal is calculated by the subtracter 4, the following calculation formula is obtained. θ) (2) = 2C sin θ The luminance signal (Y signal) can be removed to separate the color signal (C signal).

However, in the case of this method, the condition is that there is no large change in the luminance signal (Y signal) between the lines. For example, when the brightness changes between the lines, it is greatly affected. For example, in an image of a horizontal stripe in which white and black are repeated for each line, a luminance signal (Y signal) component remains from the above equation (2), and the image quality deteriorates.

(C) On the other hand, the system shown in FIG. 9 is a diagram for explaining a so-called three-dimensional YC separation system. In this example, a frame memory 7 for storing pixel information for one frame of a monitor screen, subtractors 8 and 9, a band-pass filter (BP)
F) 10. The band-pass filter (BPF) 10 is also a filter for removing noise, as described above.

In this case, 525 lines of pixel information are stored in the frame memory 7, and the subtractor 8 subtracts the current video signal from the video signal one frame before. That is, the video signal has a phase inversion of 180 ° between frames, and the subtracter 8 subtracts both signals to separate the color signal (C signal).

The creation of the luminance signal (Y signal) is the same as described above, and is obtained by subtracting the color signal (C signal) from the video signal output via the frame memory 7 by the subtractor 9.

However, even in the above-mentioned method, the condition is that there is no large change in the luminance signal (Y signal) between the respective frames. For example, when the brightness changes between the respective frames, it is greatly affected. Therefore, this method is effective mainly in processing of still images. Further, in the above-mentioned method, since the large-capacity frame memory 7 is used, the cost of the apparatus is increased.

As described above, there are various YC separation systems, and each system has a drawback. Therefore, conventionally, a YC separation system has been adopted in which the above-mentioned systems are not used independently but are used in combination.

(2) FIG. 10 is a diagram for explaining the above-mentioned combination method. In FIG. 1, reference numeral 11 denotes a YC separation unit (one-dimensional YC separation unit) using a band pass filter (BPF), and reference numeral 12 denotes a YC separation unit (two-dimensional YC separation unit) using a line memory. Reference numeral 13 denotes a correlation comparing unit which compares the horizontal correlation and the vertical correlation of the video signal. Reference numeral 14 denotes an area filter, which adjusts the correlation comparison result based on, for example, majority logic.

The switching circuit 15 includes the area filter 14
The switching process is performed according to the comparison result of the (correlation comparing unit 13), and the output of the one-dimensional YC separating unit 11 or the two-dimensional Y
The output of the C separation unit 12 is selected and supplied to a noise removal band pass filter (BPF) 16. Incidentally, reference numeral 17 denotes a delay circuit, and reference numeral 18 denotes a subtractor.

The one-dimensional YC separation section 11 has a delay circuit 11a of 1/4 of the frequency fsc of the subcarrier (subcarrier),
11b, and a one-dimensional YC separation circuit 11c. The video signal supplied via the delay circuits 11a and 11b is converted into a luminance signal (Y signal) by the one-dimensional YC separation circuit 11c.
And a color signal (C signal). On the other hand, the two-dimensional YC separation unit 12 includes the line memories 12a and 12b and the two-dimensional YC
A video signal supplied via the line memories 12a and 12b.
The signal is separated into a luminance signal (Y signal) and a chrominance signal (C signal) by c.

Each of the above-mentioned separation methods is executed by the corresponding method described above, and
One of the output of the one-dimensional YC separation circuit 11c or the output of the two-dimensional YC separation circuit 12c is selected based on the comparison result of the correlation comparison unit 13 output via the. Therefore, the chrominance signal (C signal) is extracted via the band-pass filter (BPF) 16 based on the method selected thereafter, and the luminance signal (Y signal) is subtracted by the subtracter 18 in the delay circuit 1.
7 to a color signal (C signal)
Is obtained by subtracting

It should be noted that the color signal (C
Signal) is separated into an IQ signal by a circuit (not shown),
Further, a luminance signal (Y signal) is used and separated into R (red), G (green), and B (blue) color signals by a matrix circuit.

[0021]

The above-mentioned conventional video signal separating apparatus uses NT signals having no or little phase difference in video signals.
Considering SC standard signal. For example, it is based on the assumption that the phases of the color signals of pixels separated by two lines are in phase with each other. The conventional two-dimensional YC separation circuit detects the correlation between the vertical and horizontal images based on this assumption based on the difference between the pixel values.

However, an accurate video signal is not always sent to the receiving side. For example, in the case of a reproduction signal by a VTR, the reproduction signal is greatly affected by a tape feeding speed or the like. In the case of VTR, the standard (JIS C 5581, IEC 774
According to -1,3), the tape feed speed for the video signal,
And its error is defined as V1 = 33.33 (mm / S) ± 0.5% (3).

For this reason, for example, in the case of a reproduced image of a video tape whose recording state is not good or the rotation of the motor is uneven, the pixels of each line are not always in the correct phase. In such a case, the determination of the correlation and the YC separation processing are affected. For example, FIGS. 11 and 12 are diagrams illustrating the above-described example. In both figures, ◎ indicates a target pixel, ● indicates a pixel having the same phase, ○ indicates a pixel which is 180 ° out of phase, and · indicates a pixel which is 90 ° out of phase.

First, FIG. 11 shows a state in which the sampling clock is synchronized with a subcarrier (subcarrier). For example, there is a problem in performing YC separation using a comb filter (using a two-dimensional YC separation circuit 12c). There is no.

However, if the period is shifted by 0.5% per clock for the above reason, the sampling interval is
The phase of the color signal per clock is shifted by 90 ° × 0.005 = 0.45 °.

In the video signal, the cycle of one line is defined to be 910/4 times the cycle of the sub-carrier (subcarrier). Therefore, when sampling at 4 fSC, 0.45 ° × (910/4) × 4) = 409.5 °. The problems that deviate from the NTSC standard as described above are:
This frequently occurs in a VTR reproduction signal or the like.

FIG. 12 shows an example which deviates from the NTSC standard as described above. The example shown in the figure shows a case where a 90 ° shift occurs in one line. In the case of this example, one-dimensional or two-dimensional correlation is performed by comparing the video signal one line before and the current video signal, or the current video signal and the video signal one line after, with a phase shift of 90 °. YC separation is performed.

In this case, the shift in the horizontal direction is equivalent to several clocks, so that it is negligibly small. But,
The phase is shifted in the vertical direction, and the color signal of the same color image that is originally highly correlated has the opposite phase, the correlation is reduced, and the complementary color image that is originally low in correlation is highly evaluated for the correlation.

As described above, in the conventional combined YC separation method, when the video signal does not conform to the NTSC standard, good YC separation cannot be performed. The present invention solves the above-mentioned problem by determining whether or not a video signal conforms to the standard of the NTSC standard, adopts an appropriate YC separation method, and provides a video signal separation device that does not deteriorate image quality. To provide.

[0030]

According to a first aspect of the present invention, there is provided an image processing apparatus comprising: a luminance signal (Y) included in a video signal;
In a video signal separation device that separates a color signal (C signal) from a luminance signal (Y signal) and a color signal (C signal) using a comb filter based on pixel information in the vertical direction of the screen. A vertical YC separation circuit and a horizontal Y for separating the luminance signal (Y signal) and the chrominance signal (C signal) using a band-pass filter based on pixel information in the horizontal direction of the screen.
A C separation circuit, a correlation detection circuit that detects vertical and horizontal correlations of the video signal, a determination circuit that determines whether the video signal is standard, and the determination circuit determines whether the video signal is standard. If it is determined that there is, the vertical or horizontal YC separation circuit is selected according to the output of the correlation detection circuit. If it is determined that the video signal is not standard, the output of the horizontal YC separation circuit is selected and the luminance signal (Y This can be achieved by providing a video signal separation device including a selection circuit that performs separation processing of a color signal) and a color signal (C signal).

Here, the video signal is a signal supplied from a video reproducing device such as a VTR, for example, and includes a luminance signal and 3.5.
This is a signal including a chrominance signal synthesized with a subcarrier (subcarrier) of 8 MHz, a synchronization signal, and the like.

The vertical YC separation circuit uses a comb filter to separate a luminance signal (Y signal) and a chrominance signal (C signal) for realizing a so-called two-dimensional YC separation method. This is a separation method of a luminance signal (Y signal) and a chrominance signal (C signal) that realizes a so-called one-dimensional YC separation method using a pass filter.

The correlation detecting circuit is a circuit for detecting the vertical and horizontal correlations of the video signal. For example, a subtractor is used to calculate the difference between pixels two lines apart in the vertical direction to obtain the vertical correlation. , And a difference between pixel information four pixels apart in the horizontal direction is calculated to detect a horizontal correlation.

The discrimination circuit is a circuit for discriminating whether or not the video signal is standard. For example, the discrimination circuit compares the frequency of the line clock oscillation signal with the frequency of the burst clock oscillation signal. Done by

With this configuration, if the supplied video signal is standardized in accordance with the NTSC standard, a two-dimensional YC separation system is adopted. However, NTSC is affected by the expansion and contraction of the tape and the rotation speed of the motor. If the standard is not met, a one-dimensional YC separation method having a small effect is adopted to prevent the image quality from deteriorating.

According to a third aspect of the present invention, a luminance signal (Y signal) and a chrominance signal (C signal) included in a video signal are provided.
A vertical YC separation circuit that separates the luminance signal (Y signal) and the chrominance signal (C signal) using a comb filter based on pixel information in the vertical direction of the screen; The luminance signal (Y signal) and the chrominance signal (C
Signal), and a vertical and horizontal correlation of the video signal is detected. If there is a vertical correlation, the vertical YC separation circuit is selected, and there is a horizontal correlation. A first correlation detection circuit for selecting the horizontal YC separation circuit; and a vertical YC separation circuit for detecting a color edge or a luminance edge in the horizontal direction of the screen and detecting the color edge or the luminance edge. A second correlation detection circuit for selecting a horizontal YC separation circuit when a circuit is selected and an edge of the color or an edge of luminance is not detected; a determination circuit for determining whether a video signal is standard; When the video signal is determined to be standard, the output of the first correlation detection circuit is used. When the video signal is determined to be non-standard, the output of the second correlation detection circuit is used. It can be achieved by providing a video signal separating apparatus and a road.

In the present embodiment, even in the case of an edge portion of a color or a vertical stripe luminance pattern which cannot be dealt with by the invention of the first aspect, it is determined by using the second correlation detection circuit.
A two-dimensional YC separation method is used for a color edge portion, a vertical stripe luminance pattern, and the like even in a video whose synchronization is disturbed due to expansion and contraction of a tape, fluctuation of a motor speed, and the like.

That is, in such a case, it is determined that the video signal is not standard. In this case, YC separation is performed using the output of the second correlation detection circuit, and the edge portion of the color and the luminance of the vertical stripes are determined. In the case of a pattern, a two-dimensional YC separation method is adopted.

The discriminating circuit used in this embodiment is a discriminating circuit for making a discrimination by comparing the frequency of the line clock oscillation signal with the frequency of the burst clock oscillation signal.

[0040]

Embodiments of the present invention will be described below with reference to the drawings. <First Embodiment> FIG. 1 is a system configuration diagram for explaining a video signal separating apparatus of the present embodiment.

Referring to FIG. 1, a video signal separating apparatus according to the present embodiment includes a one-dimensional YC separating section 20, a two-dimensional YC separating section 21, a correlation determining section 22, an NTSC standard determining section 23, and an AND gate (A).
ND gate) 24, switching circuit 25, delay circuit 26,
It comprises a band pass filter (BPF) 27 and a subtractor 28.

The one-dimensional YC separation section 20 includes a 1/4 fSC delay circuit (hereinafter simply referred to as a delay circuit) 29, 3
0, a one-dimensional YC separation circuit 31 and a subcarrier (subcarrier, subcarrier) using a bandpass filter (BPF).
This is a circuit for performing YC separation by removing 3.58 MHz. Here, the delay circuits 29 and 30 are circuits for delaying the pixel data output from the line memory 32, which will be described later, by １ ／ of the frequency (fSC) of the subcarrier (subcarrier). Further, the one-dimensional YC separation circuit 31 has a band pass filter (BPF) inside, and converts the video signals output from the delay circuits 29 and 30 into 3.5.
The signal of 8 MHz is removed, and the color signal (C signal) is separated.

The two-dimensional YC separation section 21 has a line memory 3
2, 33, and a two-dimensional YC separation circuit 34 for performing YC separation by comparing pixels in the vertical direction. Here, the line memories 32 and 33 each store 1
This is a memory for storing video signals for lines, and stores video signals for two consecutive lines by line memories 32 and 33. The two-dimensional YC separation circuit 34 has a subtractor therein, and subtracts a signal delayed by two lines by the line memories 32 and 33 to separate a color signal (C signal).

The correlation judging section 22 is a circuit for judging the correlation between the pixel shift in the horizontal direction and the vertical direction, and has a specific configuration shown in FIG. The correlation discrimination unit 22 is a correlation discrimination circuit 3
5. Correlation discriminating circuit 35 composed of area filter 36
Is composed of subtracters 37 and 38, an adder 39, and a comparator 40. The subtracter 37 is a subtracter for subtracting the video signal two lines before stored in the line memory 33 from a video signal newly input via a composite video input (hereinafter, simply referred to as a video input). The adder 39 calculates a value K set in advance to the subtraction result of the subtractor 37.
Are added. The value K is a value derived from past data or experimental values, for example.

The subtracter 38 is a circuit for subtracting the pixel information output from the delay circuit 30 from the pixel information output from the line memory 32. The subtraction result is output to the comparator 40. The output of the adder 39 is also supplied to the comparator 40, and the comparator 40 compares the two data. Here, the data supplied to the comparator 40 is a correlation value described later. The vertical correlation value is supplied from the subtractor 37 via the adder 39, and the horizontal correlation value is supplied from the subtracter 38. Therefore, the comparator 40 compares the magnitude of the vertical correlation value with the magnitude of the horizontal correlation value, and outputs the selection result to the area filter 36.

The area filter 36 outputs the result of the selection to the AND gate 24 according to the majority logic. The area filter 36 performs processing such as changing the separation method when the separation method is continuously changed a certain number of times so that the separation method is not frequently switched. Then, the selection signal processed by the area filter 36 is output to the AND gate (AND gate) 24.

On the other hand, the NTSC standard discriminator 23 analyzes the video signal supplied via the video input and judges whether or not the video signal matches the NTSC standard. Figure 3 shows NTS
FIG. 3 is a diagram showing a specific circuit configuration of a C standard classifier 23.
The NTSC standard discriminator 23 includes a burst lock clock oscillator 42, a line lock clock oscillator 43, a frequency comparator 44, and a low-pass filter 45.

The burst lock clock oscillator 42 extracts a sub-carrier (sub-carrier) frequency fSC (3.58 MHz) from the supplied video signal and outputs a quadrupled oscillation frequency 4 fSCk. On the other hand, the line-locked clock oscillator 43 operates at 910 during the horizontal synchronization period included in the video signal.
The frequency flck at which the clock is output is output. Normal,
The frequency fSC of the subcarrier (subcarrier) is created by a crystal oscillator or the like, and the error is very small. On the other hand, the horizontal synchronizing signal often shifts greatly due to, for example, expansion and contraction of the tape and uneven rotation of the motor.

The frequency comparator 44 compares the frequencies output from the burst lock clock oscillator 42 and the line lock clock oscillator 43, and outputs "1" when both frequencies are equal, and outputs "0" when they are different. The output of the frequency comparator 44 is output to the aforementioned AND gate (AND gate) 24 through a low-pass filter 45. The low-pass filter 45 is a circuit that suppresses a sharp change in the output of the frequency comparator 44.

The AND gate (AND gate) 24 is NT
1 according to the signal of “1” or “0” output from the SC standard discriminator 23 and the signal (signal of “1” or “0”) output from the correlation discriminator 22 (area filter 36).
Select the two-dimensional YC separation method or the two-dimensional YC separation method. For example, when the output of the NTSC standard discriminator 23 is “1”, a separation method based on the output of the correlation discriminator 22 is selected, and when the output of the NTSC standard discriminator 23 is “0”,
Select the one-dimensional YC separation method.

The output of the AND gate (AND gate) 24 is output to a switching circuit 25. The switching circuit 25 performs a one-dimensional YC separation method or a two-dimensional YC separation method in accordance with the signal output from the AND gate (AND gate) 24. Is selected, and the corresponding one-dimensional YC separation circuit 31 or two-dimensional Y
From the C separation circuit 34 to the band pass filter (BPF) 27
To output a color signal (C signal). For example, when the one-dimensional YC separation method is selected, the color signal (C signal) separated by the one-dimensional YC separation circuit 31 is output to the bandpass filter (BPF) 27 via the switching circuit 25, and the two-dimensional YC separation is performed.
When the separation method is selected, the color signal (C signal) separated by the two-dimensional YC separation circuit 34 is output to the band-pass filter (BPF) 27 via the switching circuit 25. The output of the band-pass filter (BPF) 27 is denoted by Cout.

A video signal is supplied to a subtracter 28 via a delay circuit 26. The subtracter 28 subtracts a chrominance signal (C signal) from the video signal to obtain a luminance signal (Y signal). (Y signal) is output as Yout.

In the video signal separation device having the above configuration,
The processing operation will be described below. First, a video signal reproduced by a VTR or the like is supplied to the video signal separation device of this embodiment via a video input. The video signal includes a luminance signal (Y signal), a carrier chrominance signal (C signal) combined with a subcarrier (subcarrier), a horizontal and vertical synchronization signal, and a color burst signal. Also, 1
A retrace signal is included for each line scan, and is stored in the line memory 32 for each line.

The one-line video signal stored in the line memory 32 is output to the next-stage line memory 33 and transferred to the delay circuit 29 for each pixel.
Here, the delay circuit is composed of two stages as described above,
Pixel data is held in the delay circuits 29 and 30 at a cycle of 1/4 fSC. Hereinafter, a description will be given using a specific example shown in FIG.

The example shown in FIG. 6 shows data for three lines of a video signal.
The data before the line and one line after the current line are shown. In the figure, the mark ◎ indicates a target pixel, the mark 画素 indicates a pixel which is 180 ° out of phase with respect to the target pixel, the mark 画素 indicates a pixel having the same phase as the target pixel, Marks indicate pixels that are 90 ° out of phase with respect to the target pixel. If the target pixel is the 522nd pixel of the current line (hereinafter, referred to as “522”), the pixel on the left is “521” out of phase by 90 °, and the pixel on the left is further Is “520” out of phase by 180 °. The pixel to the right of the target pixel is also “523” out of phase by 90 ° with respect to the target pixel, and the pixel to the right of the target pixel is also “524” out of phase by 180 °.
The same applies to the video signal one line before and the video signal one line after, and the pixel numbers are as shown in FIG.

Here, when the video signal shown in FIG. 4 is made to correspond to the circuit shown in FIG. 1, the following is obtained. That is,
The video signal one line before shown in FIG.
And the video signal of the current line is stored in the line memory 3.
2 is stored. The video signal after one line is in a state of being supplied via the video input. Also,
The delay circuit 30 currently holds pixel data of “520” and “521”, and the delay circuit 29 currently holds “52” and “521”.
The pixel data “2” and “523” are held.

In this state, the subtracter 37 in the correlation discriminating section 22 (correlation discriminating circuit 35) supplies the pixel “532” supplied via the video input from the pixel data “512” read from the line memory 33. Subtract the data. Further, the value K is added to the subtraction value by the adder 39. Therefore, the above formula is as follows.

| 512-532 | + K On the other hand, the subtracter 38 outputs the pixel data “520” output from the delay circuit 30 to the output of the line memory 32 (pixel data “524” (pixel data of the current line)). Is subtracted. This calculation formula is as follows.

| 520-524 | Then, the comparator 40 compares the horizontal correlation and the vertical correlation which are the above-described subtraction results, and outputs the comparison result to the area filter 36. Here, | 512-532 | + K ≦ | 520-
At the time of 524 |, it is determined that there is a correlation in the vertical direction, and for example, “1” is output to the AND gate (AND gate) 24. On the other hand, | 512-532 | + K> | 520-5
24 |, it is determined that there is a correlation in the horizontal direction, and for example, “0” is output to the AND gate (AND gate) 24. It should be noted that the value K is added to the vertical correlation side because the use of the one-dimensional YC separation method is more adaptable to a deviation such as a synchronization speed.

Therefore, at this time, a selection signal for selecting a direction having a correlation is output to the AND gate (AND gate) 24. However, even if there is a correlation, in the case of a video signal deviating from the NTSC standard, accurate YC separation cannot be performed. So, NTSC
The output of the standard classifier 23 is used.

As described above, the NTSC standard discriminator 23 has the configuration shown in FIG. 3, and the burst lock clock oscillator 42 outputs 4fSC which is four times the frequency fSC of the subcarrier (subcarrier), and outputs an accurate reference signal. It is. On the other hand, when the supply source is a VTR or the like, the signal output from the line lock clock oscillator 43 may have a problem in expansion and contraction of the tape and a driving speed of the motor as described above, and may not be able to supply an accurate synchronization signal. Many. In the case of such a video signal, the input signal to the frequency comparator 44 does not match. In this case, the signal does not conform to the NTSC standard, and accurate YC separation cannot be performed if YC separation processing is performed according to the vertical correlation.

For this reason, in this case, the frequency comparator 44
(NTSC standard discriminator 23) outputs “0”, and AND gate (AND gate) 24 outputs “0”,
Select the one-dimensional YC separation method.

On the other hand, if the signal matches the NTSC standard, the input signal to the frequency comparator 44 matches and the frequency comparator 44
(1) is output from the (NTSC standard discriminator 23), and the output of the AND gate (AND gate) 24 is made to follow the output of the correlation discriminator 22 described above. Therefore, for example, when “1” is output from the correlation discrimination unit 22, a two-dimensional YC separation method is adopted, and when “0” is output from the correlation discrimination unit 22, a one-dimensional YC separation method is adopted. .

As described above, according to the present embodiment, the VTR
For video signals that do not conform to the NTSC standard due to motor rotation errors, tape expansion or contraction, etc.
Without adopting the one-dimensional YC separation method, one-dimensional YC
The configuration adopts a separation method, and YC separation with good image quality can be performed even for a disturbed video signal such as a VTR. On the other hand, for a signal conforming to the NTSC standard, such as an on-air signal, for example, a two-dimensional YC separation method is employed to perform a good image quality YC separation process.

In the above description of the embodiment, the three-dimensional Y
Although the C separation method is not described, when a video signal that does not conform to the NTSC standard is input, a shift of one frame is accumulated.
It is also effective to apply the C separation method. <Second Embodiment> Next, a second embodiment of the present invention will be described.

FIG. 5 is a system diagram for explaining the second embodiment. Note that the same circuits as those in FIG. In this figure, this system is a one-dimensional YC
Separation unit 20, two-dimensional YC separation unit circuit 21, conventional correlation discrimination unit 22, NTSC standard discriminator 23, correlation discrimination unit 50 according to the present invention, selection circuit 51, switching circuit 25, delay circuit 26, band-pass filter ( BPF) 27 and a subtractor 28. The one-dimensional YC separation unit 20 includes delay circuits 29 and 30 and a one-dimensional YC separation circuit 31, and the two-dimensional YC separation unit 21 includes line memories 32 and 3
A three-dimensional and two-dimensional YC separation circuit 34 is provided.

The internal structure of the conventional correlation discriminating section 22 is the same as that of the first embodiment, and the description of the internal structure is omitted. The configuration of the NTSC standard discriminator 23 is the same as that described above, and includes a burst lock clock oscillator 42, a line lock clock oscillator 43, a frequency comparator 44, and a low-pass filter 45 as shown in FIG. . The difference between the present embodiment and the above-described first embodiment is that a correlation discriminating unit 50 and a selection circuit 51 according to the present invention are provided. Note that the correlation discriminating unit 5
0 and the detailed configuration of the selection circuit 51 will be described later.

In the above-described first embodiment, a one-dimensional YC separation method is employed uniformly for video signals that do not conform to the NTSC standard. However, the one-dimensional YC separation method is used for color edge portions and vertical stripe luminance patterns. Using the method causes cross-color generation. Therefore, in the present embodiment, a color edge portion and a vertical stripe luminance pattern portion are detected, and a two-dimensional YC separation method is adopted for these portions.

First, the principle of the present example will be described. Figure 1 above
As shown in FIG. 2, in a video signal that does not conform to the NTSC standard, when the video signal is sampled, the phase of the color signal (C signal) slightly shifts for each sample.

Now, as shown in the above equation (3), assuming that a deviation of 0.5% occurs between the sub-carrier (sub-carrier) and the line frequency, when sampling at 4 fSC, every sample 90 × 0.005 = 0.45 [degree / sample] (4)

For this reason, the sample after the movement of four pixels should originally be pixels having the same phase. However, there is a phase difference of 4 × 0.45 = 1.8 [degrees] between the two pixels.

Therefore, sampling of five consecutive points in a solid image of the same color is considered. Note that consecutive pixels are denoted by V ₁ , V ₂ , V ₃ , V ₄ , and V ₅ , the luminance of each pixel is denoted by Y, and the color amplitude is denoted by C. First, considering the ideal case without deviation, the sampling values of the five points are respectively

[0073]

(Equation 1)

Can be expressed as follows. And the above equation is

[0075]

(Equation 2)

Therefore, for example, taking the absolute value of the difference from the next pixel skipped by one,

[0077]

(Equation 3)

Is obtained. Where 2C | sin θ | = 2C |
cos θ |, and θ = 45 °

[0079]

(Equation 4)

Therefore, if the larger one of the two is indicated by ΔV, it is always

[0081]

(Equation 5)

The following relationship holds. therefore,

[0083]

(Equation 6)

It can be said that: On the other hand, the difference between the pixel values at points separated by 4 pixels is a shift of the color signal if the same color and the same luminance, and the shift is 1.8 ° from the equation (7). Therefore, at most 2 × sin (0.9 °) ≒ 0.0314. That is, there is a relationship of | V ₁ −V ₅ | ≦ 0.0315C (11) Substituting equation (10) into equation (11) gives

[0085]

(Equation 7)

[0086]

[0087]

(Equation 8)

Assuming that K ₁ , | V ₁ −V ₅ | ≦ K ₁ × ΔV (13) Then, when the above equation is satisfied, it can be determined that the color has changed (that is, not the same color) in that part.

With this method, a color edge is detected and Y
This example is a method of switching the C separation method. actually,
The equation (11) may be determined based on the frequency variable range of the clock generator, instead of using the error of the VTR standard to obtain the equation (11).

Further, when edge discrimination is performed only from the equation (13), a luminance vertical stripe pattern is obtained.

[0091]

(Equation 9)

In such a case, an edge cannot be determined and a cross color appears. Therefore, the difference between the pixel values in the vertical direction is further determined, and if the value is sufficiently small for several clocks, it is determined that there is no color signal (C signal), and two-dimensional YC separation is performed.

If the equation (13) is used as it is, two-dimensional YC separation is selected even in the case of color gradation, so that the image quality of a disturbed image such as a VTR deteriorates. Therefore, the predetermined value K2 is added to the right side of the above equation (13). The calculation formula in this case is as follows.

| V ₁ −V ₅ | ≦ K ₁ × △ V + K _2.
(15) 'A specific circuit configuration will be described below using the principle described above. FIG. 6 is a circuit block diagram that embodies the circuit configuration of the correlation determination unit 50 in the system configuration shown in FIG. FIG. 6 also includes the delay circuits 29 and 30 and the line memories 32 and 33 shown in FIG.

In FIG. 6, the correlation discriminating section 50 includes a correlation discriminating circuit 52a for detecting a color edge portion and a correlation discriminating circuit 52b for detecting a vertical stripe luminance pattern.
The correlation determination circuit 52a includes subtractors 53 and 54, a delay circuit 55, a comparator 56, a switching circuit 57, a multiplier 58, an adder 59, and a comparator 60. The correlation discriminating circuit 52b includes subtractors 61 and 62, comparators 63 and 64, an AND gate (AND gate) 65, four continuous delay circuits 66 to 69, and an AND gate (AND gate).
70. In addition, the correlation discriminating circuits 52a and 52
The output of 2b is output to an OR circuit 71. When at least one of the outputs of the correlation determination circuit 52a or 52b is "1" (high signal), the selection circuit 5 shown in FIG.
1 is output as "1" (high signal). That is, the correlation determination circuit 52a is a circuit that detects a color edge portion,
This is a case where a color edge is detected when "1" is output from the correlation determination circuit 52a. The correlation determination circuit 52b is a circuit for detecting a vertical stripe luminance pattern, and "1" is output from the correlation determination circuit 52b. In this case, the vertical stripe luminance pattern is detected. Therefore, when "1" is output from the OR (OR) circuit 71, this is a case where the correlation determination unit 50 detects either the color edge portion or the vertical stripe luminance pattern.

Hereinafter, a specific circuit operation will be described using the above-mentioned calculation formula. First, the circuit operation of the correlation discriminating circuit 52a for detecting a color edge will be described. The subtractor 53 is a subtractor that calculates a difference between pixel values separated by two clocks. For example, if the above-described pixels “V3” and “V4” are held in the delay circuit 29, the subtractor 53 is a line memory 3
From the next pixel “V5” supplied from the delay circuit 2
The pixel “V3” output from No. 9 is subtracted. The absolute value of this subtraction is | V3-V5 |.

The result of the subtraction by the subtractor 53 is output to the delay circuit 55 to delay the timing by one clock.
The comparator 56 compares the subtraction results at two consecutive points in time.
That is, the result of the subtraction is 1
During the clock delay, the difference in pixel value between the next two clocks is subtracted by the subtractor 53, and the result of subtraction at two successive points in time is compared by the comparator 56. That is, the pixel difference | V3-V5 | is compared with the next pixel difference | V4-V6 |.

Next, the result of the comparison by the comparator 56 is output to the switching circuit 57, and the larger one is output to the multiplier 58. Then, the value K is calculated by the multiplier 58.
1 is calculated. In addition, when the above processing result is shown by a calculation formula,
This corresponds to the right side of the above equation (13). Next, the adder 59
The above-mentioned predetermined value K2 is added. When this processing result is expressed by a calculation formula, it corresponds to the right side of the above formula (15).

On the other hand, the subtractor 54 is a subtractor that takes the difference between pixel values four clocks apart. For example, the delay circuits 29 hold the pixels “V3” and “V4” and the delay circuit 30
If the pixels “V1” and “V2” are held in
The subtractor 54 outputs “V5” supplied from the line memory 32.
Is the subtraction value of “V1” output from the delay circuit 30. When this subtraction processing is represented by an absolute value, | V1-V5
|.

This equation corresponds to the left side of the above-mentioned equation (15). The comparator 60 compares the output of the subtractor 54 with the output of the adder 59, and outputs the larger one as an OR (OR) circuit. 71. This comparison processing corresponds to the above-described equation (15), and the comparison result of the comparator 60 is the detection result of the color edge portion. In other words, in the case of a single-color solid image, even if the color signal (C signal) is shifted, the maximum value is 0.0315C (the above-described equation (11)).
By comparing the magnitude with 0, it can be determined whether the color is an edge portion of a color that changes color or is simply a shift portion of a color signal (C signal).

Therefore, according to the above equation (15),
The edge portion of the color is determined, and “1” is output to the OR (OR) circuit 71. Next, the circuit operation of the correlation determination circuit 52b for detecting the vertical stripe luminance pattern will be described. Subtractor 6
Reference numeral 1 denotes a subtracter for calculating a difference between pixel values one line away from each other. The subtracter 1 performs a subtraction process on pixel information output from the line memory 32 with corresponding pixel information one line later supplied from the video input. The subtracter 62 is also a subtractor that calculates the difference between pixel values one line apart, and subtracts the corresponding pixel information one line later from the pixel information output from the line memory 33. The comparator 63 compares the subtraction value output from the subtractor 61 with a constant value ξ, and when the output of the subtractor 61 is smaller than the constant value ξ, outputs “1” to an AND gate (AND gate) 6.
5 is output. Similarly, in the comparator 64, the subtractor 6
2 is compared with a constant value ξ.
When the output of No. 2 is smaller than the constant value ξ, “1” is output to the AND gate (AND gate) 65.

Accordingly, "1" is supplied to the AND gate (AND gate) 65 only when there is no luminance difference equal to or more than a predetermined value 垂直 in the vertical direction, and when there is no luminance difference between corresponding pixels in the vertical direction, that is, In the case of a vertical stripe luminance pattern, "1" is output.

Further, an AND gate (AND gate) 65
Are output to four-stage delay circuits 66 to 69, and the outputs of the delay circuits 66 to 69 at each stage are AND gates (A
Only when all the signals output to the ND gate 70 and supplied to the AND gate 70 are “1”, “1” is output to the OR circuit 71. That is, when the state in which the change in the luminance in the vertical direction is equal to or smaller than the constant value 連 続 continues, the image is determined to be a vertical stripe luminance pattern, and “1” is output to the OR (OR) circuit 71.

By the above circuit operation, when the supplied video signal is a color edge portion or a vertical stripe luminance pattern, the output of the OR (OR) circuit 71 becomes "1", otherwise, "0". Is output. Therefore, FIG.
When the output of the NTSC standard discriminator 23 is "1", the output of the conventional type correlation discriminator 22 is selected, and the one-dimensional YC is determined according to the vertical correlation and the horizontal correlation as in the first embodiment. The separation method or the two-dimensional YC separation method is selected. On the other hand, when the output of the NTSC standard discriminator 23 is “0”, the output of the correlation discriminator 50 is selected,
Even when a synchronization shift occurs due to a disturbance in the tape speed or the like, a two-dimensional YC separation method is used for a color edge portion or a vertical stripe luminance pattern.

With this configuration, it is possible to perform YC separation processing with excellent image quality without causing a problem of cross color even in the case of a color edge portion or a vertical stripe luminance pattern.

[0106]

As described above, according to the present invention, when the supplied video signal does not conform to the NTSC standard, the deterioration of the image quality can be prevented by adopting the one-dimensional YC separation system.

Even when the video signal is not in the NTSC standard, for example, when it is a color edge portion or a vertical stripe luminance pattern, an appropriate YC separation process is performed by employing a two-dimensional YC separation method.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram illustrating a video signal processing device according to an embodiment.

FIG. 2 is a circuit configuration diagram of a correlation determination unit.

FIG. 3 is a diagram illustrating a specific circuit configuration of an NTSC standard discriminator.

FIG. 4 is a diagram illustrating an example of three lines of pixel data included in a video signal.

FIG. 5 is a system configuration diagram illustrating a video signal processing device according to a second embodiment.

FIG. 6 is a system configuration diagram illustrating a video signal processing device according to a third embodiment.

FIG. 7 is a diagram illustrating a YC separation method using a bandpass filter (BPF) of a conventional example.

FIG. 8 is a diagram illustrating a YC separation system using a conventional comb filter.

FIG. 9 is a diagram illustrating a YC separation method using a conventional comb filter and using a memory for one frame.

FIG. 10 is a diagram illustrating a conventional example of a combination method.

FIG. 11 is a diagram illustrating a state in which a sampling example is applied to an actual video signal, and illustrates a state in which a sampling clock is synchronized with a subcarrier (subcarrier).

FIG. 12 is a diagram illustrating a state where a sampling example is applied to an actual video signal, and illustrates a state where a sampling clock is not synchronized with a subcarrier (subcarrier).

[Explanation of symbols]

 Reference Signs List 20 one-dimensional YC separation part 21 two-dimensional YC separation part 22 correlation determination part 23 NTSC standard determination part 24 AND gate (AND gate) 25 switching circuit 26 delay circuit 27 band-pass filter (BPF) 28 subtractor 29, 30 delay circuit 31 One-dimensional YC separating circuit 32, 33 Line memory 34 Two-dimensional YC separating circuit 35 Correlation discriminating circuit 36 Area filter 37, 38 Subtractor 39 Adder 40 Comparator 42 Burst lock clock oscillator 43 Line lock clock oscillator 44 Frequency comparator 45 Low pass Filter 50 Correlation determination unit 51 Selection circuit 52a, 52b Correlation determination circuit 53, 54 Subtractor 55 Delay circuit 56 Comparator 57 Switching circuit 58 Multiplier 59 Adder 60 Comparator 61, 62 Subtractor 63, 64 Comparator 65 AND gate ( ND gate) 66 to 69 delay circuit 70 AND gate (AND gate) 71 OR (OR) circuit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference)

Claims (4)

[Claims] 1. A luminance signal (Y signal) included in a video signal
And a color signal (C signal) separating the luminance signal (Y signal) and the color signal (C signal) using a comb filter based on pixel information in a screen vertical direction. A separating circuit, a horizontal YC separating circuit that separates the luminance signal (Y signal) and the chrominance signal (C signal) using a band-pass filter based on pixel information in a screen horizontal direction, and a vertical direction of the video signal, A correlation detection circuit that detects a horizontal correlation, a determination circuit that determines whether the video signal is standard, and, when the determination circuit determines that the video signal is standard, according to an output of the correlation detection circuit. When selecting a vertical or horizontal YC separation circuit and determining that the video signal is not standard, selecting an output of the horizontal YC separation circuit,
A switching processing circuit for performing a separation process of a luminance signal (Y signal) and a chrominance signal (C signal). 2. The video signal separation device according to claim 1, wherein the determination circuit performs the comparison by comparing the frequencies of a line clock oscillation signal and a burst clock oscillation signal. 3. A luminance signal (Y signal) included in a video signal.
And a color signal (C signal). A video signal separation device separates the luminance signal (Y signal) and the color signal (C signal) using a comb filter based on pixel information in a screen vertical direction. A separating circuit, a horizontal YC separating circuit that separates the luminance signal (Y signal) and the chrominance signal (C signal) using a band-pass filter based on pixel information in a screen horizontal direction, and a vertical direction of the video signal, A first correlation detection circuit that detects a horizontal correlation and selects the vertical YC separation circuit when there is a correlation in the vertical direction, and selects the horizontal YC separation circuit when there is a correlation in the horizontal direction; When detecting the edge of the color or the edge of the brightness in the direction, when detecting the edge of the color or the edge of the brightness, when selecting the vertical YC separation circuit, when not detecting the edge of the color or the edge of the brightness, A second correlation detection circuit for selecting a horizontal YC separation circuit; a discrimination circuit for discriminating whether the video signal is standard; and a first correlation for discriminating that the video signal is standard by the discrimination circuit. And a selection circuit for selecting an output of the second correlation detection circuit when using the output of the detection circuit to determine that the video signal is not standard. 4. The video signal separation device according to claim 3, wherein the determination circuit performs the comparison by comparing the frequencies of a line clock oscillation signal and a burst clock oscillation signal.