JP3393972B2 - Motion adaptive 3D Y / C separation circuit - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion adaptive three-dimensional Y / C separation circuit for separating a video signal of a television signal into a luminance signal / color signal.

[0002]

2. Description of the Related Art A video signal of a television signal is a composite of a luminance signal (hereinafter referred to as Y signal) and a color signal (hereinafter referred to as C signal) modulated by a color subcarrier, and is generally a composite video signal. Called. In the television receiver, the input composite video signal is the Y signal,
The process of separating into C signals is indispensable, and this separating process is generally called Y / C separation.

In the conventional Y / C separation circuit, it is difficult to completely separate the Y signal and the C signal, and dot interference in which the color signal leaks into the luminance signal and cross color interference in which the luminance signal leaks into the color signal occur. Had occurred. With the advance of digital technology in recent years, a so-called motion-adaptive three-dimensional Y / C separation circuit that greatly reduces these disturbances has been proposed and put into practical use.

FIG. 45 is a block diagram of a conventional motion adaptive three-dimensional Y / C separation circuit compatible with the NTSC system. Basically, the motion adaptive three-dimensional Y / C separation circuit performs inter-frame Y / C separation processing when the image is stationary (hereinafter referred to as still image), and when moving (hereinafter referred to as moving image). In-field Y / C separation processing is performed.

The composite video signal input to the input terminal 1000 has a 525H (1H is one horizontal scanning period) delay unit 1001, an adder 1002, an addition input end of a subtractor 1003, an addition input end of a subtractor 1010, Edge detection circuits 1008, 1012, two-dimensional Y / C separation circuit 100
5 is supplied.

The 525H delay unit 1001 delays the composite video signal by 525H, that is, by one frame period. The composite video signal delayed by one frame is supplied to the subtraction input terminals of the 525H delay unit 1009, the adder 1002, and the subtractor 1003.

The adder 1002 adds the input composite video signal and the composite video signal delayed by one frame to obtain an inter-frame sum signal. The one-frame sum signal is guided to the coefficient unit 1018, multiplied by 1/2, and supplied to the mixer 1014. The output of the coefficient unit 1018 is a luminance signal (hereinafter referred to as a still image Y signal) obtained during a still image.

The subtractor 1003 subtracts the composite video signal delayed by one frame from the input composite video signal to obtain a one-frame difference signal. The one-frame difference signal is a horizontal LPF (Low Pass Filter).
r) 1006, and is multiplied by 1/2 in the coefficient unit 1019 to obtain a horizontal BPF (Band Pass Fi).
lter) 1004.

The horizontal BPF 1004 has a halved 1
Band-limiting the inter-frame difference signal in the horizontal direction. Horizontal BP
The output of F1004 is a color signal obtained during a still image (hereinafter referred to as a still image C signal). This still image C signal is supplied to the mixer 1
015 is supplied.

The two-dimensional Y / C separation circuit 1005 performs Y / C separation processing for moving images. As described above, in still images, Y / C separation can be performed by calculation for one frame. This is because in the case of a still image in the NTSC system, the fact that the phase of the C signal is inverted during one frame can be used. However, moving images cannot be processed between frames because they are temporally displaced. If processing is performed for one frame, the Y signal and the C signal leak to each other, causing a failure. Therefore, the moving image is processed in the feed. Generally, Y / C separation performed only by calculation in the horizontal and vertical directions is called two-dimensional Y / C separation. FIG. 48 shows an example of the simplest two-dimensional Y / C separation.

The input terminal 1301 and the output terminal 13 shown in FIG.
07 and the output terminal 1308 are connected to the input terminal 1000, the mixer 1014, and the mixer 1015 of FIG. 45, respectively. The composite video signal input from the input terminal 1301 is supplied to the addition input end of the subtractor 1306, the 1H delay unit 1302, and the addition input end of the subtractor 1303. The 1H delay unit 1302 delays the signal by one horizontal scanning period and inputs it to the subtraction input terminal of the subtractor 1303. In the subtractor 1303, the composite video signal delayed by one horizontal scanning period from the composite video signal is subjected to subtraction processing, further multiplied by 1/2 in the coefficient unit 1304 and supplied to the horizontal BPF 1305. 1H delay device 13
02, the subtractor 1303, and the coefficient unit 1304 form a vertical bandpass filter, and perform vertical band limitation on the vertical frequency band of the color signal. The horizontal BPF 1305 is a horizontal bandpass filter centered on a color subcarrier frequency of 3.58 MHz, and its output is a color signal (hereinafter,
It is called a moving image C signal). The moving image C signal is supplied to the subtraction input terminal of the subtractor 1306 and the output terminal 1308. The moving image C signal is subtracted from the composite video signal by the subtractor 1306 to obtain a luminance signal at the time of moving image (hereinafter, referred to as moving image Y signal). This video Y signal is output from the output terminal 1
It is output to 307.

FIG. 49 shows a spectrum diagram of the moving picture Y signal and the moving picture C signal separated by the two-dimensional Y / C separation circuit configured as described above. The shaded C portion is the moving image C signal, and the other Y portion is the moving image Y signal.

Returning to FIG. 45, the description will be continued. Mixer 101
Reference numeral 4 outputs a luminance signal (hereinafter referred to as a mixed Y signal) in which a still image Y signal and a moving image Y signal are mixed according to a motion signal described later. The mixer 1015 performs the same processing as that of the mixer 1014 and outputs a color signal (hereinafter referred to as a mixed C signal) in which the still image C signal and the moving image C signal are mixed.

Next, a motion detecting process for generating a motion signal for controlling the mixers 1014 and 1015 will be described.

There are two processes for detecting the motion, that is, one-frame motion detection for detecting one-frame motion and two-frame motion detection for detecting two-frame motion. Both have their advantages and disadvantages and are usually designed to make up for their shortcomings.

First, the one-frame motion detection will be described.

A horizontal LPF 1006 extracts a horizontal low frequency component from the one-frame difference signal output from the subtractor 1003. Since the one-frame difference signal includes a still image C signal and a motion component between one frames, it is necessary to remove the still image C signal in order to detect the motion between one frames. The horizontal low-frequency component of the one-frame difference signal from which the still image C signal has been removed is supplied to the one-frame motion detection circuit 1007.

The motion detection circuit 1007 for one frame is set to 1
The horizontal low-frequency component of the inter-frame difference signal is converted into an absolute value, and when the difference is a certain value or more, it is considered that there is motion, and a motion signal is generated. The amount of motion of one frame of the input composite video signal is detected, and the motion signal of one frame is detected by the MAX circuit 1.
Supply to 013. In addition, the motion detection circuit 10 for one frame
07, the edge detection circuit 1008 switches the sensitivity of motion detection.

The edge detection circuit 1008 detects an edge component from the input composite video signal, determines whether it is an edge, and switches the sensitivity of the one-frame motion detection circuit 1007.

The internal configuration and operation of the one-frame motion detection circuit 1007 and the edge detection circuit 1008 are the same as those of the two-frame motion detection circuit 1011 and the edge detection circuit 1012 which will be described later, and only the input signals are different. Detailed description is omitted here.

Next, the motion detection between two frames will be described. The composite video signal delayed by one frame period output from the 525H delay unit 1001 is 525H.
The delay unit 1009 further delays by one frame period. Therefore, the 525H delay unit 1009 outputs a composite video signal delayed by two frame periods with respect to the input composite video signal. The composite video signal delayed by two frames is subtracted by the subtractor 1010.
Is supplied to the subtraction input terminal of.

The subtractor 1010 subtracts the composite video signal delayed by two frames from the composite video signal and supplies the difference between the two frames to the motion detection circuit 1011 between the two frames. The difference signal between two frames is a motion component between two frames of the Y signal and the C signal. The inter-two-frame motion detection circuit 1011 detects the amount of motion between the two frames and supplies a two-frame motion signal to the MAX circuit 1013. The motion detection circuit 1011 between two frames is
The edge detection circuit 1012 switches the sensitivity for detecting the motion amount between the edge portion and the flat portion of the input image.

Detailed operations of the inter-frame motion detection circuit 1011 and the edge detection circuit 1012 will be described with reference to FIG.

The edge detection circuit 1012 uses the horizontal BPF 1
101, an absolute value circuit 1102, a comparator 1103, and a pixel enlargement circuit 1104. The two-frame motion detection circuit 1011 includes an absolute value circuit 1105, nonlinear processing circuits 1106 and 1107, and a switch 1108. 46, the input terminal 1100, the input terminal 1109,
The output terminals 1110 are the input terminals 10 in FIG.
00, output terminal of subtractor 1010, MAX circuit 1013
It is connected to the.

First, the edge detection circuit 1012 extracts a horizontal high frequency component from the input composite video signal by the horizontal BPF 1101. The horizontal BPF 1101 is, for example, a filter having a center frequency of 3.58 MHz. Horizontal B
The horizontal high frequency component extracted by the PF 1101 is subjected to absolute value conversion by the absolute value circuit 1102 and supplied to the comparator 1103. The comparator 1103 makes a comparison with the threshold value ref1 and determines that there is a horizontal edge when the value is ref1 or more. The comparison result of the comparator 1103 is the pixel enlarger 1
The signal extended by 104 is supplied to the inter-frame motion detection circuit 1011 as an edge determination signal.

The two-frame difference signal input from the input terminal 1109 is guided to the two-frame motion detection circuit 1011, converted into an absolute value by the absolute value circuit 1105, and supplied to the non-linear processing circuits 1106 and 1107. .

Non-linear processing circuits 1106 and 1107 perform non-linear processing on the absolute value-converted two-frame difference signal, and output a motion signal of several bits. Non-linear processing circuit 1
The non-linear processing circuit 106 and the non-linear processing circuit 1107 are non-linear processing circuits for a portion other than the edge portion (hereinafter referred to as a flat portion) and an edge portion, respectively, and have different characteristics. An example of the characteristic diagram of the nonlinear processing circuits 1106 and 1107 is shown in FIG.
Shown in.

In FIG. 47, the horizontal axis represents the value of the input signal,
The vertical axis represents the value of the output signal. Here, the nonlinear processing circuit 1
Motion signals output from 106 and 1107 are 3 bits.
And The solid line shows the characteristic of the non-linear processing circuit 1106, and the dotted line shows the characteristic of the non-linear processing circuit 1107.

Now, the relationship between the mixers 1014 and 1015 and the motion signal used for control will be described. Since the characteristics of the processing for still images and the processing for movies are very different,
For example, if the motion signal is controlled as a binary signal, the moment of switching becomes unnatural visually. Therefore, when the motion is small, the unnaturalness is reduced by mixing the still image processed signal and the moving image processed signal.

In FIG. 47, in the flat part of the image,
The characteristic shown by the solid line is taken, and the output becomes 0 until the input level is a1, and the mixers 1014 and 1015 in FIG. 45 output only the signals subjected to the still image processing. Until the input level exceeds a1 and reaches a2, the output is 1, 2, 3, ..., 6
The ratio of moving image processing will increase compared to still image processing. When the input becomes a2 or more, the output becomes 7, and only the video processed signal is output.

In FIG. 47, the edge portion of the image has the characteristic indicated by the dotted line, which is a characteristic that is easier to obtain a still image than the characteristic of the flat portion. This is to reduce the influence of noise or the like, and to prevent the image from being determined as a moving image although it is actually a still image.

Returning to FIG. 46, the description will be continued. Switch 11
08 selects the output of the non-linear processing circuit 1106 in the flat part and the output of the non-linear processing circuit 1107 in the edge part by the output of the edge detection circuit 1012.

As described above, the inter-two-frame motion detection circuit 1011 outputs the inter-two-frame motion signal indicating the amount of motion between the two frames.

Returning to FIG. 45, the description will be continued. The 1-frame motion detection circuit 1007 outputs a 1-frame motion signal indicating the motion amount between 1 frames, and the 2-frame motion detection circuit 1011 outputs a 2-frame motion signal indicating the motion amount between 2 frames. Each is output to the MAX circuit 1013.

The MAX circuit 1013 selects the larger one of the one-frame motion signal and the two-frame motion signal and outputs it to the mixers 1014 and 1015. The larger one is selected because both have advantages and disadvantages and reduce omission of motion detection.

Here, since the motion detection for one frame is performed based on the difference between one frame, it is possible to detect the motion of one frame cycle (15 Hz), for example, but the phase in which the C signal is multiplexed is the frame. Because it is inverted every time,
Even if there is no movement in the image, the part with color will be judged as a moving image. Therefore, the determination is made only on the horizontal low-frequency component where the C signal component is low. Therefore, when the horizontal high-frequency component of the Y signal is moved, or when only the color signal is changed, it cannot be detected.

On the other hand, in the motion detection between two frames, in the case of a still image between two frames, the phase of the C signal is the same,
When the movement of the horizontal high frequency component of the Y signal occurs, it is possible to detect even when only the color changes. However, a motion of one frame cycle (15 Hz) cannot be detected because no difference occurs.

Therefore, the MAX circuit 1013 selects the larger one of the one-frame motion signal and the two-frame motion signal to control the mixers 1014 and 1015.
The omission of motion detection is reduced.

The outputs of the mixers 1014 and 1015 are output to the output terminal 1016 as a mixed Y signal and to the output terminal 1017 as a mixed C signal, respectively.

[0040]

In the conventional motion adaptive three-dimensional Y / C separation circuit configured as described above, it is possible to completely separate the Y signal and the C signal for a still image. It was possible to eliminate dot interference in which the signal leaks into the Y signal and cross color interference in which the Y signal leaks into the C signal. However, in the moving image, since the Y / C separation is performed with the spectrum as shown in FIG. 49, it is impossible to suppress the occurrence of the interference. In particular, with respect to cross color interference, colors different from the original ones are generated, so they are very noticeable and the image quality is remarkably deteriorated.

As a method of reducing cross color interference, there is a simple method of lowering the sensitivity of motion detection. Hereinafter, a case where the sensitivity of motion detection is lowered will be described. In FIG. 50, the solid line is the characteristic of the one-frame difference calculation used to detect the one-frame motion and the still image color signal, and the dotted line is the characteristic of the two-frame difference calculation used to detect the two-frame motion. , The alternate long and short dash line is the characteristic of the one-frame sum operation. In FIG. 45, the output is from the coefficient unit 1019 whose gain is ½, but here, the adder 1002,
The outputs of the subtracters 1003 and 1010 are normalized and considered to be 1. The horizontal axis represents frequency in the time direction, and the vertical axis represents gain. When the input composite video signal is completely stationary, the Y signal is 0 Hz and the C signal is 15H in the time direction.
in z.

Now, it is assumed that the Y signal has a movement of aHz. The output of the difference between 1 frames is b1, and the output of the difference between 2 frames is b2. If the sensitivity of the motion detection between 1 frames or the motion detection between 2 frames is lowered and the C signal is subjected to the still image processing in spite of the motion, the Y signal leaks into the C signal by b1. Colors will be added to the areas that are not originally colored. However, since the gain is low,
It is relatively lower than the cross color interference generated by moving image processing.

Further, comparing the one-frame motion detection with the two-frame motion detection shows a great effect when the sensitivity of the two-frame motion detection is lowered. In the one-frame motion detection, only the motion of the horizontal low frequency component can be detected as described in the conventional example. Therefore, the movement of the Y signal in the shaded portion in FIG. 49, which causes the cross color interference due to the moving image processing, cannot be detected.

As described above, it is possible to reduce cross-color interference by lowering the sensitivity of motion detection between two frames, but the performance of motion detection, which is the original purpose, is degraded and various interference occurs. To do. If the sensitivity of motion detection between two frames is lowered, still image processing will be performed if only the color changes. An example of the problem will be described with reference to FIG. FIG. 51
Shows a case where a circular red figure moves from A to B on a blue screen. It is assumed that there is no level difference between the Y signals of the blue and red parts. If the sensitivity of motion detection between two frames is low and the entire still image processing is performed, interference occurs at the shaded portion. In the processing of the C signal, a color in which blue and red are mixed is obtained, and in the processing of the Y signal, the C signal leaks into the one-frame sum component, so that dot-like interference occurs.

The present invention has been proposed to solve the above problem, and it is a motion adaptive three-dimensional Y / C separation in which the occurrence of cross color interference is reduced without lowering the performance of motion detection. The purpose is to provide a circuit.

[0046]

[Means for Solving the Problems]

(First Configuration Example) In a motion adaptive three-dimensional Y / C separation circuit that separates a composite video signal into a luminance signal and a chrominance signal in accordance with the amount of motion of an image, a first field from the composite video signal is processed by processing within one field First color signal extracting means for extracting the color signal of the composite video signal, second color signal extracting means for extracting the second color signal from the composite video signal by processing for one frame, and one frame of the composite video signal. First motion detecting means for detecting a motion amount of a video from a difference component between the two, and first motion detecting means for extracting a first motion detection signal; and detecting a motion amount of the video from a difference component between two frames of the composite video signal, A second motion detecting means for extracting a second motion detecting signal; a mixing means for mixing the first color signal and the second color signal; 1-frame sum extraction means for extracting a sum signal between frames, and edge detection for detecting a horizontal edge component from the 1-frame sum signal and generating an edge detection signal indicating whether or not it is larger than a first predetermined value. Means for inputting the second motion detection signal and the second predetermined value, and when the edge detection signal indicates that the edge detection signal is larger than the first predetermined value, the second predetermined value is selected, and otherwise , The selecting means for selecting the second motion detection signal, and the first maximum value means for outputting the output of the selecting means and the signal of the larger level of the first motion detection signal. The mixing operation of the mixing means is controlled by the output of the first maximum value means.

(Second Configuration Example) In a motion adaptive three-dimensional Y / C separation circuit which separates a composite video signal into a luminance signal and a chrominance signal according to the amount of motion of an image, the composite video signal is processed by processing in one field. First color signal extraction means for extracting a first color signal from the composite video signal, second color signal extraction means for extracting a second color signal from the composite video signal by processing for one frame, The first motion detecting means for detecting the motion amount of the image from the difference component between the 1 frames and extracting the first motion detection signal and the motion amount of the image from the difference component between the 2 frames of the composite video signal. Second motion detection means for detecting and extracting a second motion detection signal; mixing means for mixing the first color signal and the second color signal; and the composite video signal. A one-frame sum extraction means for extracting a one-frame sum signal from the signal; a one-frame difference extraction means for extracting a one-frame difference signal from the composite video signal; Edge detecting means for generating an edge detection signal from a first edge signal detecting an edge component and a second edge signal detecting a vertical low frequency component of a horizontal edge component from the one frame difference signal; A second motion detection signal and a first predetermined value are input, and in the edge detection signal, the first edge signal is larger than a second predetermined value and the second edge signal is lower than a third predetermined value. When indicating that the first
Selecting a predetermined value, and selecting the second motion detection signal when the edge detection signal indicates any other value, a larger one of the output of the selection means and the first motion detection signal Maximum value means for outputting a level signal, and the mixing operation of the mixing means is controlled by the output of the maximum value means.

(Third configuration example) In a motion adaptive three-dimensional Y / C separation circuit for separating a composite video signal into a luminance signal and a chrominance signal according to the amount of motion of the image, the composite video signal is processed by processing within one field. A first color signal extracting means for extracting a first color signal from the composite video signal; a second color signal extracting means for extracting a second color signal from the composite video signal by processing for one frame; First motion detecting means for detecting a motion amount of a video from a difference component between 1 frames and extracting a first motion detection signal, and detecting a motion amount of a video from a difference component between 2 frames of the composite video signal. And a second motion detecting means for extracting a second motion detecting signal, a mixing means for mixing the first color signal and the second color signal, and the composite video. 1-frame sum-of-frames extraction means for extracting a 1-frame sum-of-frames signal from the signal, and horizontal edge components are detected from the 1-frames sum-of-frames signal to detect whether or not it is larger than a first predetermined value Edge detection means for generating an edge detection signal indicating the presence or absence of an edge, first control means for controlling the first predetermined value from the first color signal, the second motion detection signal, and When a predetermined value of 2 is input and the edge detection signal indicates that the edge detection signal is larger than the first predetermined value, the second predetermined value is selected, and when the other values are indicated, the second motion detection signal is selected. Selecting means for selecting, an output of the selecting means and the first motion signal,
Maximum value means for outputting a signal of a level having a larger movement amount, and the mixing operation of the mixing means is controlled by the output of the maximum value means.

(Fourth Configuration Example) In a motion adaptive three-dimensional Y / C separation circuit for separating a composite video signal into a luminance signal and a chrominance signal in accordance with the amount of motion of an image, the composite video signal is processed by processing within one field. First color signal extraction means for extracting a first color signal from the composite video signal, second color signal extraction means for extracting a second color signal from the composite video signal by processing for one frame, The first motion detecting means for detecting the motion amount of the image from the difference component between the 1 frames and extracting the first motion detection signal and the motion amount of the image from the difference component between the 2 frames of the composite video signal. Second motion detection means for detecting and extracting a second motion detection signal; mixing means for mixing the first color signal and the second color signal; and the composite video signal. An inter-frame sum extraction unit that extracts an inter-frame sum signal from the signal, and a horizontal edge component is detected from the one-frame sum signal to detect whether or not it is larger than a first predetermined value. Edge detection means for generating an edge detection signal indicating the presence or absence of an edge, control means for controlling the first predetermined value from the first color signal and the composite video signal, and the second movement. The second predetermined value is selected when the detection signal and the second predetermined value are input and the edge detection signal indicates that the edge detection signal is larger than the first predetermined value, and the second movement is selected otherwise. The mixing operation of the mixing means includes a selection means for selecting a detection signal, and a maximum value means for outputting the signal of the output of the selection means and the level of the larger motion amount of the first motion signals. Is the maximum value means Characterized in that it is controlled by the output.

[0050]

[0051]

DETAILED DESCRIPTION OF THE INVENTION

[0052]

【Example】

(First Embodiment) FIG. 1 shows a first embodiment of the present invention. The present embodiment is a modification of the conventional example shown in FIG. 45, and those having the same function are designated by the same reference numerals and the description thereof will be omitted. A portion surrounded by a dotted line in FIG. 1 is a portion added by the present invention. The processing of the Y signal is exactly the same as in the conventional example of FIG.
Signal processing is being changed.

The output of the motion detection circuit 1011 between two frames is sent to the MAX circuit 102 via the edge control circuit 101.
To one end of the input of the
The output of 07 is led to the input of the other end of the MAX circuit 102. The MAX circuit 102 is functionally the same as the MAX circuit 1013, and the maximum value of two inputs is input to the mixer 101.
Output to 5.

The internal structure of the edge control circuit 101 is shown in FIG.
Shown in. The input terminal 201 of FIG. 2 is the adder 100 of FIG.
2 is connected to the output terminal of FIG. 2 and the input terminal 206 of FIG.
2 is connected to the output terminal of the inter-frame motion detection circuit 1011 and the output terminal 208 of FIG. 2 is connected to one end of the input terminal of the MAX circuit 102 of FIG.

The one-frame sum signal input from the input terminal 201 is input to the input terminal A of the comparator 204 via the horizontal BPF 202 and the absolute value circuit 203. The comparator 204 receives the constant value ref input from the input terminal B.
3 is compared, and when the signal input from the input terminal A is large, a high level (hereinafter, referred to as H level) signal is output, and in other cases, a low level (hereinafter, referred to as L level) signal is output. To do. The output of the comparator 204 is input to the pixel enlargement circuit 205.

On the other hand, the signal input from the input terminal 206 is the output of the inter-frame motion detection circuit 1011 in FIG. 1, and is input to one end of the input terminal of the selector 207. 0 is input to the other end of the input terminal of the selector 207, and the output is output to the output terminal 208. The output terminal 208 is connected to one end of the input terminal of the MAX circuit 102 in FIG.

Now, the basic concept of the present invention will be described.
In the motion adaptive three-dimensional Y / C separation circuit shown in the conventional example, when it is determined that there is a motion, both the Y signal and the C signal are two-dimensional Y.
/ C separation was performed. However, focusing only on the C signal, the signals are multiplexed around a part of the band (fsc), and in a general natural image or the like, the energy tends to decrease as the distance from the center increases. Therefore, it is intended to reduce the cross color interference generated by the moving image processing by performing the still image processing on the visually inconspicuous portion even if the portion is moving.

FIG. 4 shows the relationship between the Y signal and the C signal. fsc becomes the DC component of the C signal when demodulated after Y / C separation. In other words, it is a plane portion, and if still images are processed even if they are moving, the colors are mixed, so that they are easily noticeable. The portion indicated by the dotted line is the horizontal edge portion of the C signal, and is hard to be noticed if the area subjected to the still image processing is small. Therefore, the present invention does not reduce the cross color generated by the Y signal in the vicinity of fsc, but reduces the cross color interference generated when the Y signal is present at the position corresponding to the horizontal edge portion of the C signal. I am trying.

Returning to FIG. 2, the operation of the circuit will be described. The characteristic of the horizontal BPF202 is 3.58MHz as in the conventional example.
For example, the filter is not centered in the vicinity but in the vicinity of 1.79 MHz as shown in FIG. When the characteristic of the horizontal BPF 202 is too broad, 3.
Since the gain in the vicinity of 58 MHz becomes high, the vicinity of the DC component of the C signal is also extracted. Therefore, horizontal B
It is desirable that the PF 202 be composed of a filter having a second derivative or higher. FIG. 5 shows the relationship between the output of the horizontal BPF 202, the absolute value circuit 203, the comparator 204, and the pixel enlargement circuit 205 with respect to the edge of the Y signal which is now stationary. Here, the horizontal BPF 202 is a second-order differential filter. The pixel enlargement circuit 205 is inserted in order to prevent the portion a of the output of the comparator 204 in FIG. FIG. 7 shows an example of the configuration of the pixel enlargement circuit 205.

In FIG. 7, an input terminal 701 and an output terminal 707 serve as the input and output of the pixel enlargement circuit 205. The signal input from the input terminal 701 is guided to the 1-clock delay unit 702 and the MAX circuit 706. The 1-clock delay devices 702, 703, 704, and 705 are delay devices each performing a delay of 1 clock, and are connected in cascade, and the respective outputs are guided to the MAX circuit 706. M
The AX circuit 706 outputs the maximum value of the five inputs to the output terminal 70.
Output to 7. Therefore, in this example, the 1-clock delay unit 70
This means that the output of 3 is centered and the pixels are stretched by about 2 pixels before and after.

Here, when the output of the pixel enlargement circuit 205 is at the H level, the selector 207 selects 0. Therefore, no matter what value the two-frame motion signal input from the input terminal 206 has, the selector 207 outputs 0. This means that even if the 2-frame motion detection circuit 1011 of FIG. 1 determines that there is motion, if the 1-frame motion detection circuit 1007 does not determine that it is motion, the processing of the C signal is performed as a still image. Nothing else.

The frequency of the Y signal in the time axis direction is 15
If the still image processing is mistakenly performed in the case of Hz, the result shown in FIG.
The characteristic of the difference between 1 frames shown by the solid line of 0 has the maximum gain, and the level at which the Y signal is processed as the C signal is also high. However, the characteristic of the one-frame sum input from the input terminal 201 is that the gain is 0 when the frequency in the time axis direction is 15 Hz, as shown by the alternate long and short dash line in FIG.
Therefore, the selector 207 of FIG.
Select 06. That is, the output of the motion detection circuit 1011 between two frames in FIG. 1 is output as it is, and is processed as a moving image.

As described above, it is possible to provide a motion adaptive three-dimensional Y / C separation circuit in which cross color interference is reduced without degrading the motion detection performance.

(Second Embodiment) The second embodiment is an improvement of the first embodiment. For example, 7.5 in the time direction.
When a Y signal having a Hz component is input, both the characteristics of the one-frame sum and the one-frame difference in FIG. 50 have relatively high gains. Therefore, in rare cases, the motion detection for two frames may stop, and the interference in which the Y signal leaks into the C signal may be noticeable. The second embodiment aims to reduce the above-mentioned slight interference while reducing the cross-color interference.

Here, the basic concept of the second embodiment will be described. In the first embodiment, of the Y signals that cause cross color interference, the motion detection between the two frames in the band in which the still image processing can be performed is stopped. In the second embodiment, the region in which the motion detection between the two frames is stopped is limited to only the portion where the cross color interference is large, so that the slight leakage of the Y signal into the C signal as described above is reduced. To do.

The second embodiment is a modification of the first embodiment. The changed part is only the internal configuration of the edge control circuit 101 in FIG. FIG. 6 shows the internal configuration of the edge control circuit 101 which is the second embodiment of the present invention. Therefore, of the constituent elements described in FIG. 2, those having the same function are denoted by the same reference numerals, and description thereof will be omitted.

The inter-frame sum signal input from the input terminal 201 is input to the horizontal BPFs 202 and 601.
The configuration and operation from the horizontal BPF 202 to the pixel enlargement circuit 205 are the same as those in the first embodiment, and the description will be omitted. The horizontal BPF 601 is a first-order differential bandpass filter having a peak gain in the same band as the horizontal BPF 202, and has, for example, the characteristics shown by the solid line in FIG.
The output of the horizontal BPF 601 is converted into an absolute value by the absolute value circuit 602 and input to the input terminal A of the comparator 603. A constant value ref4 is input to the input terminal B, and the comparator 603 outputs an H level signal when the value input to the input terminal A is larger than the value of the input terminal B, and an L level signal otherwise. The signal is output to the pixel compression circuit 604.

FIG. 15 shows a horizontal BPF 601, an absolute value circuit 602, and a comparator 6 for the edge of a stationary Y signal.
03, the relationship of the output of the pixel compression circuit 604 is shown.

As described in FIG. 5 of the first embodiment,
The output of the comparator 204 in FIG. 5D is prevented by the pixel expansion circuit 205 in order to prevent the portion a from coming off.
As shown in (e), a few pixels are expanded and output.

Therefore, since the H level period of FIG. 5E is only for still image processing, when the Y signal leaks into the C signal, there is a case in which the period during which the signal is disturbed is widened and conspicuous.

Therefore, the pixel compression circuit 60 shown in FIG.
Generate a signal that shows the central part of the edge like 4 outputs,
Only when this signal is in the H period, the output of the pixel stretching circuit 205 is enabled, and the period in which the motion detection between two frames is stopped is narrowed.

An example of the configuration of the pixel compression circuit 604 is shown in FIG.
Shown in. In FIG. 8, an input terminal 801 and an output terminal 80
5 is an input / output of the pixel compression circuit 604. Input terminal 80
The signal input from 1 is a 1-clock delay unit 802, M
It is led to the IN circuit 804. 1-clock delay unit 802,
Reference numeral 803 is a delay device that cascades delays for one clock, and the outputs of the delay devices 803 are guided to the MIN circuit 804. The MIN circuit 804 outputs the minimum value of the three inputs to the output terminal 805. Therefore, in this example, the output of the 1-clock delay unit 802 is reduced by about 1 pixel before and after.

Returning to FIG. 6, the description will be continued. AND the output of the pixel expansion circuit 205 and the output of the pixel compression circuit 604
The product is obtained by the circuit 605, and the selector 207 is controlled by the output of the AND circuit 605.

As described above, in the second embodiment,
As compared with the first embodiment, the area where the motion detection between two frames is stopped is narrowed in the horizontal direction and is limited to the central portion of the edge in the horizontal direction. The portion where the cross color is largely generated when the moving image is processed is the central portion of the edge in the horizontal direction, and the cross color reduction effect does not decrease. On the other hand, since the still image processing is performed, even when the Y signal leaks into the C signal, the area in which interference occurs is narrower than in the first embodiment, so that it is less noticeable.

It should be noted that the simplest method for narrowing the region where the motion detection between two frames is stopped in the horizontal direction is as shown in FIG.
The pixel compression circuit 604 of FIG. 6 may be inserted between the output of the pixel expansion circuit 205 and the control terminal of the selector 207, but the configuration shown in this embodiment can be narrowed optimally. It is possible.

Further, in FIG. 6, the pixel compression circuit 60
Needless to say, even if the number 4 is omitted, it is possible to narrow the width in the horizontal direction. This is because in a filter having a gain peak in the same band, the transient response period is shorter in the first derivative than in the second derivative.

(Third Embodiment) The third embodiment is also an improvement of the first embodiment. Here, the basic concept of the third embodiment will be described. In the first and second embodiments,
Among the Y signals that cause cross color interference, the motion detection between the two frames in the band where there is no problem even when the still image processing is performed is stopped. The band is a region of a portion in FIG. 12, and is a horizontal frequency band of a horizontal edge portion of the C signal.
As shown in FIG. 12, band limitation is not performed in the vertical direction.

Therefore, in the third embodiment, as shown in part b of FIG. 13, by slightly limiting the band in which the motion detection between two frames is stopped to the vertical high band, a slight occurrence occurs in the case of still image processing. It reduces the interference.

The reason why the vertical high range is limited is that the energy of the DC component is generally high in a natural image, and tends to decrease with increasing high range.

The simplest means for limiting the band in the vertical high range is to insert a vertical BPF as shown in FIG. 9 between the horizontal BPF 202 and the absolute value circuit 203 of FIG. 2 in the first embodiment. For example, the characteristics are such that the gain peaks at 240 LPH and the gain becomes 0 at 0 LPH and 480 LPH. However, in order to configure the vertical BPF, a line memory is required for delaying by 1H, and a capacity corresponding to the number of output bits of the horizontal BPF 202 is required.

Therefore, in the third embodiment, the vertical band limitation in which the capacity of the line memory is suppressed to the necessary minimum is realized to suppress the increase in cost due to the increase in hardware. The third embodiment will be described below.

The third embodiment is a modification of the first embodiment. The changed part is only the internal configuration of the edge control circuit 101 in FIG. FIG. 16 shows the internal configuration of the edge control circuit 101 which is the third embodiment of the present invention. Therefore, of the constituent elements described in FIG. 2, those having the same function are denoted by the same reference numerals, and description thereof will be omitted.

The inter-frame sum signal is input from the input terminal 201 and supplied to the horizontal BPF 202 and the horizontal BPF 2301. The configuration and operation from the horizontal BPF 202 to the comparator 204 are the same as those of the first embodiment shown in FIG.

The horizontal BPF 2301 is a bandpass filter having the same first derivative as the horizontal BPF 601 of FIG. 6 of the second embodiment, and its characteristics are also the same as those of FIG. The code determination circuit 2302 detects the polarity of the output of the horizontal BPF 2301. The actual code determination operation is, for example, the horizontal BPF2.
If the output of 301 is a signed binary representation having positive / negative values, its most significant bit (sign bit) is simply extracted. The output of the code determination circuit 2302 is supplied to the normalization circuit 2303.

Here, the horizontal BPF 2301, the code determination circuit 2302, the comparator 204, and the normalization circuit 230 are shown in FIG.
3 shows the relationship of the outputs of FIG.

It is now assumed that the inter-frame sum signal input to the input terminal 201 has a rising edge on the n-th line and a falling edge on the (n + 1) -th line as shown in FIG. The output of the horizontal BPF 2301 is as shown in (a) because it is a first-order differential, and the output of the code determination circuit 2302 is as shown in (c).
It becomes H level when the output of the PF 2301 has a negative polarity. On the other hand, the output of the comparator 204 is as shown in (D) as described in the first embodiment.

The normalization circuit 2303 includes a code determination circuit 23.
02 output and the output of the comparator 204 are normalized so that the arithmetic processing of the vertical BPF 2304 described later can be performed.
First, the output of the code determination circuit 2302 is set to the sign bit and added to the output of the comparator 204 to form a 2-bit configuration. The relationship at that time is shown in FIG. This 2-bit signal represents three states in four values. That is, since there are a plurality of states without edges, the normalization circuit 230
In 3, the three states are normalized so as to be represented by three values, as shown in FIG.

FIG. 18 shows an example of a specific configuration of the normalization circuit 2303. Through the input terminal 2501, the output of the sign determination circuit 2302 is supplied to one input terminal of the AND circuit 2502, and the output of the comparator 204 is supplied to the other input terminal of the AND circuit 2502 and the output terminal 2503. When there is no edge, the output of the comparator 204 becomes L level and the AND 2502 outputs L level regardless of the output of the code determination circuit 2302. Further, when there is an edge, the output of the comparator 204 becomes H level, so the code determination circuit 23
02 output becomes valid. The output of the AND circuit 2502 is converted into a sign bit and output from the output end 2503. Therefore, when the input value at the input end 2501 is -2, the output value at the output end 2503 is normalized to 0.

Returning to FIG. 16, the description will be continued. Normalization circuit 2
The output of 303 is supplied to the vertical BPF 2304, and the vertical high frequency component is extracted. FIG. 9 shows an example of the configuration of the vertical BPF.

The output of the normalization circuit 2303 is the input terminal 901.
Is supplied to the addition input of the 1H delay unit 902 and the subtractor 903 via. The output of the normalization circuit 2303 delayed by 1H by the 1H delay unit 902 is supplied to the subtraction input of the subtractor 903 to form a 2-tap vertical bandpass filter. The output 904 outputs the subtractor 903 output. Since the output of the normalization circuit 2303 has a 3-bit value with 2 bits, the output of the vertical BPF 2304 has 3 bits.
It has a 5-valued value as shown in 7.

As can be seen from FIG. 27, the normalization circuit 23
If the 03 output does not change in the vertical direction, it is removed and becomes 0. It has a value other than 0 when it changes in the vertical direction. Since the output of the normalization circuit 2303 is 2 bits, 1H delay 902
The capacity of the line memory of is only 2 bits.

Returning to FIG. 16, the description will be continued. Vertical BPF2
The output of 304 is supplied to the comparator 2305. The comparator 2305 is an equal comparator that determines whether it is 0 or a value other than 0. That is, the vertical BPF 2304
When the output of is 0, the comparator 2305 outputs L level, and when it is other than 0, it outputs H level. Comparator 230
The output of 5 is supplied to the pixel enlargement circuit 2306.
The pixel enlargement circuit 2306 performs the same operation as the pixel enlargement circuit 205 of FIG. 2 of the first embodiment. The output of the pixel enlargement circuit 2306 is guided to the control input terminal of the selector 207 and controls the selector 207.

As described above, in the third embodiment, in addition to the first embodiment, the band for stopping the motion detection between two frames is further limited to the vertical high frequency component. Therefore, it is possible to reduce the interference caused by the Y signal leaking into the C signal when the still image processing is performed, although it is slight. Further, since the bit number of the line memory used for the vertical BPF is only 2 bits, the increase in hardware scale can be suppressed to the minimum.

It is of course possible to combine the third embodiment with the second embodiment. The third embodiment, the second
19 shows an example of a block diagram when applied to the embodiment of FIG. The horizontal BPF 2301 of FIG. 16 of the third embodiment and the horizontal BPF 601 of FIG. 6 of the second embodiment have the same characteristics 1.
Since it is a bandpass filter of the second derivative, it can be shared.
Therefore, in the second embodiment of FIG. 6, the code determination circuit 2302,
Normalization circuit 2303, vertical BPF 2304, comparator 23
A configuration having both the advantages of the second and third embodiments can be realized only by adding 05.

Another example of combining the second and third embodiments is the configuration example of FIG. In this example, the output of the comparator 603 is connected to the code determination circuit 2302 and the normalization circuit 2.
This is an example in which a 303, a vertical BPF 2304, and a comparator 2305 are added. Also in this example, after the output of the code determination circuit 2302 is added to the output of the comparator 603 and normalized, the vertical B
By performing the PF processing, vertical band limitation is applied. Figure 1
9 and FIG. 20, the output result of the AND circuit 605 is almost the same.

(Fourth Embodiment) In the fourth embodiment, the second,
As in the case of the third embodiment, the first embodiment is improved, and the interference that occurs slightly is reduced by limiting the band in which the motion detection between two frames is stopped.

Fourthly, the basic concept of the embodiment will be described.

In the third embodiment, the band in which the motion detection between two frames is stopped is limited to the vertical high band. In the fourth embodiment, the motion detection between two frames is normally operated when the vertical low-frequency component when the Y signal moves is high. Specifically, 1 having a peak at 15 [Hz]
A horizontal high frequency component and a vertical low frequency component are extracted from the inter-frame difference signal, and when the level is high, the control for stopping the motion detection between the two frames is stopped. The operation area of the fourth embodiment is the same as the second and third embodiments of the fourth embodiment shown in FIG. 21, except that a part of the first embodiment is changed. The changed part is shown in Figure 1.
Only the internal configuration of the edge control circuit 101 of FIG. Figure 10
In addition, the edge control circuit 101 according to the fourth embodiment of the present invention.
The internal structure of is shown. Of the constituent elements described in FIG. 2 of the first embodiment, FIG. 6 of the second embodiment, and FIG. 16 of the third embodiment, those having the same function will be described with the same reference numerals added. Is omitted.

The one-frame difference signal output from the subtractor 1003 in FIG. 1 is passed through the input end 1701 and the horizontal BPF1.
702, and is supplied to the horizontal BPF 2301. Horizontal BPF
The steps from 202 to the selector 207 are the same as those in the first embodiment shown in FIG. 2, and the horizontal BPF 2301 to the normalization circuit 2303.
The configurations and operations up to this point are the same as those in the third embodiment except that the input signal is changed from the one-frame sum signal to the one-frame difference signal, and the description thereof is omitted.

The horizontal BPF 1702 is the horizontal BPF 202.
4 is a bandpass filter having the same characteristics as in FIG. Horizontal B
The horizontal high frequency component corresponding to the horizontal edge component extracted by the PF 1702 is converted into an absolute value by the absolute value circuit 1703, and the comparator 1704 compares it with the constant value ref5. Constant value ref
When it is larger than 5, an H level signal is output. Third
In the same manner as in the embodiment of FIG.
ign bit, and the output of the comparator 1704 and the data are combined to form a 2-bit configuration.
Supply to. After being normalized to three values by the normalization circuit 2303, it is supplied to the vertical LPF 1705.

FIG. 11 shows an example of the configuration of the vertical LPF 1705.
Shown in . The output of the normalization circuit 2303 is supplied to the 1H delay unit 1802 and the adder 1803 via the input end 1801. The output of the normalization circuit 2303 delayed by 1H by the 1H delay unit 1802 is supplied to the other input terminal of the adder 1803. Therefore, the 1H delay device 1802 and the adder 1
803 forms a 2-tap vertical low-pass filter.

Here, as in the third embodiment, since the output of the normalization circuit 2303 has 2 bits and 3 values, the output of the vertical LPF 1705 is 3 bits and has 5 values as shown in FIG. Has a value of. If there is a horizontal edge component and there is no change in the vertical direction, the output value of the vertical LPF 1705 becomes a value of 2 or -2. Since the output of the normalization circuit 2303 is 2 bits, the capacity of the line memory of the 1H delay unit 1802 is 2 bits.

The output of the vertical LPF 1705 is the output terminal 18
It is supplied to the absolute value circuit 1706 of FIG.
The signal whose absolute value has been converted by the absolute value circuit 1706 is supplied to the comparator 1707. The comparator 1707 is an equal comparator that outputs an H level signal when the input signal has the same value as the value to be compared. The value to be compared is 2, and an H level signal is output when there is a horizontal edge component and there is a change in the vertical direction.

The output of the comparator 1707 is supplied to the pixel enlargement circuit 1708. The configuration and operation of the pixel enlargement circuit 1708 are the same as those of the pixel enlargement circuit 205.
The output of the pixel enlargement circuit 1708 is inverted by the inverter 1709 and supplied to the AND circuit 1710. AND
The circuit 1710 calculates the product of the pixel enlargement circuit 205 and the pixel enlargement circuit 1708, and supplies the product to the control input terminal of the selector 207.

Therefore, when the level of the vertical low frequency component of the horizontal edge of the one-frame difference signal is high, the output of the pixel stretching circuit 205 is ignored, and the motion adaptive processing is performed according to the result of the normal two-frame motion detection. Done.

The features of this embodiment will be described with reference to FIG. . FIG. 22 is an enlarged version of FIG. 50 described in the first embodiment up to 15 Hz on the time axis.

Now, it is assumed that the input signal is only the Y signal and the signal is moving at the time frequency aHz. In this case, as described in the first embodiment, even if the color signal is subjected to the static image processing, the interference due to the leakage of the Y signal to the C signal due to the one-frame difference characteristic is b1 and is inconspicuous.

However, there is a movement of time frequency cHz,
When the level of the Y signal is extremely high, the gain is d3 due to the inter-frame sum characteristic, but the comparison value r of the comparator 204 is r.
The level of ef3 may be exceeded. In such a case, in the first and second embodiments, the level of the Y signal leaking into the still image color signal generated from the difference between the frames becomes d1 and the interference becomes noticeable. In such a case, in the present embodiment, the control for stopping the motion detection between the two frames is stopped by the output of the inverter 1709, the normal motion detection between the two frames is operated, and the C signal for the moving image is output.

Further, it is of course possible to combine with the second embodiment. FIG. 23 shows a configuration diagram of an application example in which the second embodiment and the fourth embodiment are combined. This can be easily realized by adding a circuit from the horizontal BPF 1702 of the fourth embodiment to the AND circuit 1710 to the second embodiment and supplying the output of the AND circuit 605 to one input of the AND circuit 1710.

(Fifth Embodiment) A fifth embodiment will be described. FIG. 24 shows a block diagram of the fifth embodiment. Fifth
This embodiment is a partial modification of FIG. 1 of the first embodiment. Therefore, of the constituent elements described in FIG. 1, those having the same function are designated by the same reference numerals, and description thereof will be omitted.

The internal configuration of the edge control circuit 101 shown in FIG. 24 of the fifth embodiment is applicable to any of the first to fourth embodiments described above.

The difference between FIG. 24 and FIG. 1 is that the coefficient unit 101
That is, the vertical BPF 3101 is inserted between the No. 9 and the horizontal BPF 1004. The one-frame difference signal output from the coefficient unit 1019 is subjected to vertical band limitation by the vertical BPF 3101. Vertical BPF3101 is 240LPH
The vertical bandpass filter has a characteristic that the gain peaks at 0 and the gain becomes 0 at 0LPH and 480LPH. The output of the vertical BPF 3101 is supplied to the horizontal BPF 1004 and subjected to horizontal band limitation.

With the above configuration, it is possible to reduce the leakage of the Y signal to the C signal, which is slightly generated by stopping the 2-frame motion detection. Even if the Y signal leaks into the C signal, this is the vertical low-frequency component of the inter-frame difference signal processed as a color signal (vertical high-frequency component of the color when demodulated).
Is reduced.

In this embodiment, the problem of the motion adaptive three-dimensional Y / C separation circuit can be improved.

The characteristics of the 1-frame difference and 2-frame difference used for the 1-frame motion detection and 2-frame motion detection have the characteristics shown in FIG. When moving at a frequency of 30 Hz, the gain is 0 in both cases, so it is impossible to detect the movement. Therefore, when the Y signal has a time frequency of 30 Hz and is in the horizontal band of the color signal, the still image is processed even though it is a moving image, and the Y signal leaks into the C signal. By limiting the vertical band of the C signal for still image as in this embodiment, it is possible to reduce the leakage of the vertical low-frequency component of the Y signal.

Generally, the C signal is visually less sensitive than the Y signal. Therefore, even if the band is limited as in the present embodiment, the image quality is hardly deteriorated.

Although the vertical BPF 3101 is inserted between the coefficient unit 1019 and the horizontal BPF 1004 in this embodiment, it goes without saying that the same effect can be obtained by inserting it between the horizontal BPF 1004 and the mixer 1015.

In addition to the above embodiments, the present invention can be applied to a system corresponding to the second-generation EDTV broadcasting which has already been put into practical use. Second-generation EDTV broadcasting is currently N
It is compatible with TSC broadcasting, and various reinforcement signals for improving image quality are multiplexed. The reinforcement signal (hereinafter referred to as the horizontal reinforcement signal) used to reproduce the horizontal high frequency component of 4.2 [MHz] or more is multiplexed in a band having a conjugate relationship with the C signal when a still image is received. When reproducing on the side, a motion adaptive three-dimensional Y / C separation circuit is required. in this case,
The difference from the first to fifth embodiments is that the signal obtained by subtracting the horizontal reinforcement signal from the one-frame difference signal by the space-time filter is used only for processing the C signal for still image. Therefore, it is possible to easily apply the present invention to reduce cross color interference.

(Sixth Embodiment) The sixth embodiment is an improvement of the second embodiment. In the second embodiment, cross color interference that occurs when the Y signal is present in the band corresponding to the horizontal edge of the C signal is reduced.

Here, in order to further reduce the cross color interference, the constant value ref3 of the comparator 204 and the constant value ref4 of the comparator 603 in FIG. 6 are set to small values. When the constant values ref3 and ref4 are reduced, the comparator 20
Even when the value of the A terminal of 4 is small, the still image processing is performed, and as a result, the horizontal band of the still image processing is widened.

Assuming that the horizontal BPFs 202 and 601 in the preceding stage of the constant values ref3 and ref4 have the characteristics shown in FIG. This is because the available bandwidth is expanded.

This means that even a horizontal edge portion of a comparatively gentle C signal moves, and a still image may be processed when a one-frame sum component occurs. In such a case, which is a very rare case, the C signal is mixed in a relatively large area, and the C signal is noticeable as a blur of the C signal to deteriorate the image quality.

In this embodiment, as described above, the constant value ref
By lowering the values of 3 and ref4, it is intended to further reduce the cross color interference, and also to reduce the bleeding of the C signal, which rarely occurs at that time.

Here, the basic concept of the sixth embodiment will be described. First, to further reduce cross-color interference,
The constant values ref3 and ref4 are set low. At that time, if there is a horizontal edge of the C signal having a high degree of saturation, ref3 and ref4 are set to a higher value so that the C signal does not bleed. However, when the horizontal edge of the C signal has a movement of 15 Hz, all the bands in the time direction, the vertical direction, and the horizontal direction are the same as the band in which the still image processing is performed to suppress the occurrence of cross color interference, I can't judge. Therefore, the characteristics of the signal in a portion slightly away from the pixel on which the still image processing is performed are examined, and when the saturation of the C signal is high, the constant values ref3 and ref4 are controlled to be higher.

Although the above-mentioned contents are the basic concept, in the present embodiment, the constant values ref3 and ref are set due to the circuit configuration.
It is not necessary to control 4 simultaneously. Edge control circuit 600
Similarly to the edge control circuit 101, 1 is an AND circuit 605.
Then, the output of the pixel expansion circuit 205 and the pixel compression circuit 6
The output of 04 is taken.

Therefore, the value input to the B terminal of the comparator 603, which affects the output of the pixel compression circuit 604, determines the output value of the edge control circuit 6001. That is, the output of the edge control circuit 6001 can be controlled only by setting the constant value ref4 of the B terminal of the comparator 603.

There is also a meaning of saving hardware, and in this embodiment, a method of controlling the area for still image processing by setting only the constant value ref4 will be described.

FIG. 29 shows the configuration of the sixth embodiment. The sixth embodiment is similar to the second embodiment in FIG. 1 of the first embodiment.
It is a part of the change. Therefore, of the constituent elements described in FIG. 1, those having the same function are assigned the same reference numerals and description thereof will be omitted. In FIG. 29, the portion surrounded by the dotted line is the portion changed in this embodiment.

The moving picture C signal, which is one output of the two-dimensional Y / C separation circuit 1005, is led to one end of the input of the edge control circuit 6001 via the color saturation detection circuit 6000.
The color saturation detection circuit 6000 pays attention to the peripheral pixels of the portion corresponding to the horizontal edge of the C signal and continuously outputs the H level signal when the signal level is higher than a certain fixed value, and the L level in other cases. The signal of is output. FIG. 30 shows the internal structure of the color saturation detection circuit 6000.

The input terminal 3701 has a two-dimensional Y of FIG.
The moving image C signal which is one output of the / C separation circuit 1005 is input, and the output terminal 3711 is guided to the edge control circuit 6001 in FIG. The moving image C signal input from the input terminal 3701 is sent to the comparator 3 via the absolute value circuit 3702.
Input to the input terminal A of 703. A constant value ref37 is input to the input terminal B of the comparator 3703. The comparator 3703 outputs an H level signal when the value of the input terminal A is larger than the value of the input terminal B, and outputs an L level signal in other cases.

2ck delay device 3704, 2ck delay device 37
05 and 2ck delay units 3706 are connected in cascade, and the output of the comparator 3703 is 2ck delay units 3704 and 2ck.
It is led to one of the input terminals of the AND circuit 3708 via the delay unit 3705, the 2ck delay unit 3706, and the 2ck delay unit 3707.

Each 2ck delay unit 3704, 3705, 37
The outputs of 06, 3707 and the comparator 3703 are A
The product is taken by the ND circuit 3708 and is led to one input terminal of the OR circuit 3710, and at the same time, the 1ck delay unit 370
9 is led to the other input terminal of the OR circuit 3710.

The OR circuit 3710 includes an AND circuit 370.
The output of 8 and the output of the 1 ck delay unit 3709 are summed and output to the output terminal 3711.

FIG. 31 shows an absolute value circuit 3702, a comparator 3703, and a 1 ck delay unit 37 for a signal having a high color saturation.
09, the relationship of the output of the OR circuit 3710 is shown. here,
A signal having a high color saturation will be described on the assumption that the horizontal frequency of fsc, which is DC of the C signal, is 3.58 MHz.

The color saturation detection circuit 6000 has the input terminal 3
When a signal such as that shown in the actual battle of (A) is input to 701, the absolute value is converted by the absolute value circuit 3702, and the output of the absolute value circuit 3702 becomes as shown in (A). 31 shown in FIG.
The mark indicates a sampling point.

The output (b) of the absolute value circuit 3702 is input to the A terminal of the comparator 3703 and compared with the constant value ref37 input to the B terminal, and the value input to the A terminal is
When it is larger than the constant value ref37, an H level signal is output, and in other cases, an L level signal is output.

Since the sampling point corresponds to the portion where the system clock (c) changes from the L level to the H level, the output of the comparator 3703 is as shown in (d). Therefore, when the output signal of the comparator 3703 is continuously at the H level, the 2ck delay unit 3704, the 2ck delay unit 37
05, 2ck delay device 3706, each output of 2ck delay device 3707 becomes H level, and the output of AND circuit 3708 also becomes a signal of H level.

Now, the roles of the last-stage 1ck delay unit 3709 and the OR circuit 3710 shown in FIG. 30 will be described. As shown by the broken line in (A) of FIG. 31, a signal whose phase is shifted compared to the waveform shown in the solid line of (A) is input terminal 3701.
In the case of being input to the output of the comparator 3703, the output of the comparator 3703 becomes a signal such as (K) in which the H level and the L level are interchanged with respect to the signal of (D). Then, the AND circuit 37
The output of 08 is an L level signal.

That is, depending on the phase of the signal input to the input terminal 3701, the color saturation detection circuit 6000 should originally be used.
However, even though the H level signal is output, the detection omission that outputs the L level signal occurs. Therefore, the OR circuit 3710 plays the role of preventing the detection omission by taking the sum of the value of the signal 1 ck before and the value of the current signal.

FIG. 32 shows the configuration of the edge control circuit 6001. The edge control circuit 6001 changes only the part related to the comparator 603 of the edge control circuit 101 of FIG. 6 used in the first embodiment. Therefore, of the constituent elements described in FIG. 6, those having the same function are denoted by the same reference numerals and description thereof will be omitted.

The output of the absolute value circuit 602 is the comparator 603.
Input to the A terminal of. The output of the selector 3901 is guided to the B terminal of the comparator 603. The constant value ref4 is input to the L-side input terminal of the selector 3901, and the constant value ref4 ′ is input to the H-side input terminal. Note that r
The relationship between ef4 and ref4 ′ is ref4 <ref4 ′.

The output signal of the color saturation detection circuit 6000 is input to the input terminal 3900, and the output signal of the selector 3901 is controlled. When the signal input to the input terminal 3900 is in the H level period, the selector 3901 outputs ref4 ′ to the input terminal B of the comparator 603.

When the signal input to the input terminal 3900 is in the L level period, the selector 3901 outputs ref4 to the input terminal B of the comparator 603.

In the horizontal edge portion of the slightly moving C signal as shown in FIG. 34, when the color saturation of the peripheral pixels is high, the color saturation detection circuit 6000 outputs an H level signal to the edge control circuit 6001. Then, the selector 3901 selects ref4 ′ having a relatively large value.

On the other hand, when the color saturation of the peripheral pixels of the horizontal edge portion of the C signal is low, the color saturation detection circuit 6000 outputs an L level signal to the edge control circuit 6001 and the selector 3901 has a relatively small size. Select the value ref4.

Therefore, when the selector 3901 selects ref4 ′, the band for still image processing becomes narrower, and when the selector 3901 selects ref4, the band for still image processing becomes wider.

That is, by controlling the area for still image processing in accordance with the detection result of the color saturation detection circuit 6000, cross color interference can be further reduced, and the bleeding of the C signal, which occurs very rarely at that time. It can be reduced.

In the present embodiment, the saturation level of the C signal is detected for one side, for example, the left side (L) area with the horizontal edge of the C signal as a boundary shown in FIG. 34. It is also possible to detect the saturation of the C signal in the right side (R) region, which is the other side, and control the region for still image processing.

(Seventh Embodiment) In the sixth embodiment, a region considered to have a high color saturation is detected from the C signal before demodulation, and the comparison value ref4 in FIG. 6 is controlled. In the seventh embodiment, the constant value ref4 is controlled by using the I and Q signals obtained by demodulating the C signal. If I and Q signals are used,
It becomes easy to detect a change in hue.

Therefore, in this embodiment, the degree of saturation is detected and the portion where the hue changes in the horizontal direction, that is, the horizontal edge portion of the color signal is detected, and only when both conditions are met, the condition of FIG. The comparison value ref4 is controlled to be set higher.

FIG. 33 shows the configuration of the seventh embodiment.
Only an improvement is made to FIG. 6 of the second embodiment. Therefore, of the constituent elements described in FIG. 6, those having the same function are denoted by the same reference numerals and description thereof will be omitted. The part surrounded by the dotted line in FIG. 33 is the part improved in this embodiment.

The detection method using the I signal and the Q signal shown in this embodiment detects that pixels with a high signal level are continuous and the hue is changed.

That is, as shown in FIG. 35, in consecutive pixels with the horizontal edge of the C signal as a boundary, (1) the level of the I signal or the Q signal is not less than a certain value, and (2) I
When the absolute value of a-Ib or the absolute value of Qa-Qb is a certain value or more, and when both of the above two points are satisfied,
It is configured to output an H level signal.

The moving picture C signal, which is one output of the two-dimensional Y / C separation circuit 1005, is led to one end of the input of the edge control circuit 6001 via the color saturation detection circuit 7000.
The color saturation detection circuit 7000 pays attention to the peripheral pixels of the portion corresponding to the horizontal edge of the C signal, and the above-mentioned (1),
When the condition of (2) is satisfied, an H level signal is output, and in other cases, an L level signal is output. FIG. 36 shows the internal structure of the color saturation detection circuit 7000.

The moving picture C signal input to the input terminal 4000 is demodulated into an I signal and a Q signal by the color demodulation circuit 4001.
The I signal is subjected to absolute value conversion in the absolute value circuit 4002, and is guided to the hue detection circuit 4003 and the continuity detection circuit 4004.

In the hue detection circuit 4003, the absolute value of Ia-Ib shown in FIG.
Detects whether it is above a certain value, and if it is above a certain value,
An H level signal is output, otherwise an L level signal is output.

The demodulated Q signal has an absolute value circuit 4011.
Is converted into an absolute value and is guided to the hue detection circuit 4012 and the continuity circuit 4013.

In the hue detection circuit 4012, the absolute values of Qa-Qb shown in FIG.
Detects whether it is above a certain value, and if it is above a certain value,
An H level signal is output, otherwise an L level signal is output.

In the OR circuit 4009, the hue detection circuit 400
By taking the sum of the output of 3 and the output of the hue detection circuit 4012, the above-mentioned (2) is detected.

The output of the absolute value circuit 4002 is guided to the continuity detection circuit 4004, and it is detected whether or not the level of the I signal is a certain value or more in several consecutive pixels. H level signal, otherwise,
It outputs an L level signal.

Further, the output of the absolute value circuit 4011 is guided to the input of the continuity detecting circuit 4013, and it is detected whether or not the Q signal level is a certain value or more in several consecutive pixels, and the certain value In the case of, an H level signal is output, and in other cases, an L level signal is output.

In the OR circuit 4006, the continuity detection circuit 40
The output of 04 and the output of the continuity detection circuit 4013 are summed to obtain the I on one side with the horizontal edge of the C signal as a boundary.
The signal level of the signal or the Q signal is detected.

The output of the continuity detecting circuit 4004 is guided to the input of the delay circuit 4005 and delayed by several clocks.
Similarly, the output of the continuity detecting circuit 4013 is also introduced to the input of the delay circuit 4014 and delayed by several clocks.
The OR circuit 4008 sums the output of the delay circuit 4005 and the output of the delay circuit 4014 to detect the signal level of the I signal or the Q signal on the other side with the horizontal edge of the C signal as a boundary.

The AND circuit 4007 replaces the OR circuit 4006.
And the output of the OR circuit 4008 are taken. And A
In the ND circuit 4010, the output of the OR circuit 4009 and the AN
By taking the product of the outputs of the D circuit 4007, both the above conditions (1) and (2) can be satisfied.

Here, the internal structure of the hue detection circuit 4003 is shown in FIG. The output of the absolute value circuit 4002 input to the input terminal 400 is the input for addition of the subtractor 401,
It is led to the input for subtraction of the subtractor 401 via the delay device 402.

The output of the subtractor 401 is the absolute value circuit 403.
Is converted into an absolute value and input to the A terminal of the comparator 404. A constant value ref70 is input to the B terminal,
If the value input to the A terminal of the comparator 404 is larger than the value of the B terminal, an H level signal is output.
Output level signal.

1ck delay device 405, 1ck delay device 40
6, the 1 ck delay unit 407 and the 1 ck delay unit 408 are connected in cascade, and the output of the comparator 404 is the 1 ck delay unit 4
05, 1ck delay device 406, 1ck delay device 407, 1c
It is led to one of the input terminals of the AND circuit 409 via the k delay device 408. The output of the comparator 404 and the outputs of the 1 ck delay units 405, 406, 407 and 408 are multiplied by the AND circuit 409 and output to the output terminal 410.

Since the internal structure of the hue detecting circuit 4012 is exactly the same as that of the hue detecting circuit 4003, the description thereof will be omitted.

Next, FIG. 38 shows the internal structure of the continuity detecting circuit 4004. The output of the absolute value circuit 4002 input to the input terminal 411 is input to the A terminal of the comparator 412. A constant value ref71 is input to the B terminal, and an H level signal is output when the value input to the A terminal of the comparator 412 is larger than the B terminal value, and an L level signal otherwise. Is output.

1ck delay device 413, 1ck delay device 41
4, the 1 ck delay unit 415 and the 1 ck delay unit 416 are connected in cascade, and the output of the comparator 412 is the 1 ck delay unit 4
13, 1ck delay device 414, 1ck delay device 415, 1c
It is led to one of the input terminals of the AND circuit 417 via the k delay device 416. The output of the comparator 412 and the outputs of the 1 ck delay units 413, 414, 415 and 416 are multiplied by the AND circuit 417 and output to the output terminal 418. In addition,
The continuity detection circuit 4013 has the same internal configuration as that of the continuity detection circuit 4004, and therefore description thereof will be omitted.

In the sixth embodiment, the constant value ref4 in FIG. 6 is controlled only by the saturation of the C signal. For example, in FIG.
It is assumed that the saturation degree of the portion (L) of is detected. Figure 44
When the color saturation of the portion a is high and the luminance signal that causes the cross color interference is present in the portion b as shown in FIG. 6, the comparison value ref4 in FIG. 6 is set to a high value in the portion b. , The effect of reducing cross color is reduced. on the other hand,
In the seventh embodiment, since the set value is increased only in the portion where the hue changes in the horizontal direction, the comparison value r
Since ef4 is set to be low, the occurrence of cross color can be suppressed to a sufficiently low level.

(Eighth Embodiment) In the eighth embodiment, the sixth
Further improvements are made to the embodiment of. In the sixth embodiment, when the surrounding color saturation is high, the constant value ref of FIG.
4 is set higher. Therefore, the cross color reduction performance deteriorates in an image having a high color saturation as a whole and including an edge of a luminance signal that causes cross color interference when moving image processing is performed. Therefore, in the present embodiment, even if the color saturation of the peripheral portion is high, if the level of the area considered to be the edge of the luminance signal is high, the setting value ref4 is set low to sufficiently reduce the cross color interference. Makes it possible.

FIG. 39 shows the structure of the eighth embodiment. The eighth embodiment is obtained by modifying a part of FIG. 29 of the sixth embodiment. Therefore, of the constituent elements described in FIG. 29, those having the same function are denoted by the same reference numerals and description thereof will be omitted. The part surrounded by the dotted line in FIG. 39 is the part improved in this embodiment.

The composite video signal is guided to the pixel number detection circuit 8002 via the horizontal edge detection circuit 8000. Further, it is guided to the pixel number detection circuit 8002 via the vertical edge detection circuit 8001.

The output of the pixel number detection circuit 8002 is ANDed.
The output of the color saturation detection circuit 6000 is guided to one input of the circuit 8003, and is guided to the other input of the AND circuit 8003.

The output of the AND circuit 8003 is led to the edge control circuit 6001. FIG. 41 shows the internal configuration of the horizontal edge detection circuit 8000, the vertical edge detection circuit 8001, and the pixel number detection circuit 8002.

The composite video signal input to the input terminal 4300 is passed through the vertical LPF 4301 and the horizontal BP.
It is guided to F4302 to detect the horizontal edge of the Y signal.

The composite video signal input to the input terminal 4300 is guided to the horizontal LPF 4318 via the vertical BPF 4317 and detects the vertical edge of the Y signal.

The output of the horizontal BPF 4302 is subjected to absolute value conversion in the absolute value circuit 4303 and input to the A terminal of the comparator 4304. A constant value ref80 is input to the B terminal, and the value input to the A terminal of the comparator 4304 is B
When it is larger than the value of the terminal, an H level signal is output, and in other cases, an L level signal is output to one input terminal of the NOR circuit 4305.

Similarly, the output of the horizontal LPF 4318 is converted into an absolute value by the absolute value circuit 4319, and the comparator 43
It is input to the A terminal of 20. A constant value ref is applied to the B terminal.
81 is input, and when the value input to the A terminal of the comparator 4320 is larger than the value of the B terminal, an H level signal is output, and otherwise, an L level signal is output.
Output to the other input terminal of 305.

The output of the NOR circuit 4305 is guided to the 1ck delay unit 4306. 1ck delay device 4306, 1ck
Delay device 4307, 1ck delay device 4308, 1ck delay device 4309, 1ck delay device 4310, 1ck delay device 431
The 1 and 1 ck delay units 4312 and 1 ck delay units 43013 are connected in cascade, and the output of the NOR circuit 4305 is guided to the 1 ck delay unit 4306 as described above. NOR
The output of the circuit 4305 is 1ck delay device 4306, 1ck
Delay device 4307, 1ck delay device 4308, 1ck delay device 4309, 1ck delay device 4310, 1ck delay device 431
It is guided to one of the input terminals of the adder 4314 via the 1 and 1 ck delay unit 4312 and the 1 ck delay unit 43013.

The output of the NOR circuit 4305 and the outputs of the 1ck delay circuits 4306 to 4313 are added by the adder 4314.
And added to the A terminal of the comparator 4315.

A constant value re is applied to the B terminal of the comparator 4315.
When f82 is input and the value input to the A terminal of the comparator 4315 is larger than the value of the input terminal B, an H level signal is output to the output terminal 4316 otherwise. . That is, the circuit shown in FIG. 41 detects the number of pixels that do not correspond to the horizontal edge and the vertical edge of the Y signal.

As described above, in the eighth embodiment, as in the sixth embodiment, peripheral pixels have high color saturation and correspond to the horizontal and vertical edges of the Y signal of the peripheral pixels. Only when the number of pixels is small, the comparison value re of FIG.
Set f4 higher.

Therefore, in the image including the edge of the Y signal, which has high color saturation as a whole and cross color interference occurs when moving image processing is performed, the comparison value ref4 is set to a low value and sufficient cross clock is generated. It is possible to obtain a color reduction effect.

(Ninth Embodiment) The ninth embodiment can be adapted to the first to eighth embodiments described so far, and the performance can be further improved. Here, a case where it is applied to the second embodiment will be described.

In the first to eighth embodiments, the cross-color interference is reduced by forcibly processing the motion detection between two frames of the C signal as a still image by the signal generated from the one-frame sum signal. There is.

The reason why the one-frame sum signal is used has been described in the first embodiment. Here, Y having a large level is used.
When the horizontal edge portion of the signal quickly moves between frames, the edge component one frame before may be processed as a still image up to a region that should be processed as a moving image.

For example, if the input terminal 1000 of FIG.
When a composite video signal including an edge component as shown in the signal 0 of (0) is input and motion occurs in one frame, the output of the 525H delay unit 1001 becomes as shown in (i). The output of the adder 1002, that is, the one-frame sum signal is as shown in FIG. 40C, and is input to the edge control circuit 101. The operation of the edge control circuit 101 has already been described in the first embodiment and will not be repeated here.

The signal shown in FIG. 40C is the input terminal 2
01, through the horizontal BPF 202, as shown in (d).
The signal shown in (d) corresponds to the absolute value circuit 203 and the comparator 20.
4. Through the pixel enlargement circuit 205, as shown in (e), the detection result is that the edge positions are points a and b, even though the edge position at the present time is point b.

Also, in the case of passing through the horizontal BPF 601, the absolute value circuit 602, the comparator 603, and the pixel compression circuit 604 of FIG. 6, there are two detection points as in the above case. As a result, a region that should be processed as a moving image is processed as a still image.

Although the above operation rarely causes the deterioration of the image quality, the comparison value ref4 of FIG. 6 is normally set low as in the sixth, seventh and eighth embodiments. If the horizontal edge portion of the C signal with high saturation moves quickly, colors may be mixed.

With respect to the current signal, there is no problem because the value of ref4 is set higher by the C signal separated by the two-dimensional Y / C separation, but this is not applied to the signal one frame before. .

Here, the basic outline of the ninth embodiment will be described. In the present embodiment, in order to solve the above problem, the edge control circuit 101 also performs processing by adding an edge generated from a composite video signal indicating the information of the current signal.

FIG. 42 shows the configuration of the ninth embodiment. The ninth embodiment is a modification of part of FIG. 1 used in the second embodiment. Therefore, of the constituent elements described in FIG. 1, those having the same function are assigned the same reference numerals and description thereof will be omitted. The part surrounded by the dotted line in FIG. 42 is the part improved in this embodiment.

The composite video signal input to the input terminal 1000 is guided to the edge control circuit 9000. FIG. 43 shows the internal structure of the edge control circuit 9000. However, the internal configuration of the edge control circuit 9000 is obtained by modifying a part of FIG. 6 of the second embodiment. Therefore, of the constituent elements described in FIG. 6, those having the same function are denoted by the same reference numerals and description thereof will be omitted.

The composite video signal input to the input terminal 901 is subjected to absolute value conversion by the absolute value circuit 903 via the horizontal BPF 902, and is introduced to the A terminal of the comparator 904. A constant value ref9 is input to the B terminal of the comparator 904. When the value input to the comparator 904 is larger than the value of the input terminal B, an H level signal is output, and otherwise, an L level signal is input. The signal is output to the pixel expansion circuit 905.

The outputs of the pixel expansion circuit 905, the pixel expansion circuit 205, and the pixel compression circuit 604 are connected to the AND circuit 9 respectively.
The product is obtained at 06 and the selector 207 is controlled. In addition,
The horizontal BPF 902 and the pixel enlargement circuit 905 perform the same operations as the horizontal BPF 202 and the pixel enlargement circuit 205, respectively, and therefore description thereof will be omitted.

As described above, if the horizontal edge of the composite video signal is used, it is possible to eliminate the area where the still image is processed by the edge one frame before, which is a very rare case. Such deterioration of image quality can be reduced.

[0200]

As described above, according to the motion adaptive three-dimensional Y / C separation circuit of the present invention, it is possible to greatly reduce cross color interference without degrading the motion detection performance.

[Brief description of drawings]

FIG. 1 is a motion adaptive three-dimensional Y / C according to a first embodiment of the present invention.
It is a figure which shows the structure of a separation circuit.

FIG. 2 is an edge control circuit 101 according to the first embodiment of the present invention.
FIG.

3 is a characteristic diagram of a horizontal BPF of the edge control circuit 101 of FIG.

FIG. 4 is a spectrum diagram of a television signal for explaining the operation of the present invention.

FIG. 5 is an edge control circuit 101 according to the first embodiment of the present invention.
6 is a waveform diagram for explaining the operation of FIG.

FIG. 6 is an edge control circuit 101 according to a second embodiment of the present invention.
It is a figure which shows the structure of.

FIG. 7 is a circuit diagram showing a specific example of a pixel stretching circuit 205 of the edge control circuit 101 according to the first and second embodiments of the present invention.

FIG. 8 is an edge control circuit 101 according to a second embodiment of the present invention.
6 is a circuit diagram showing a specific example of the pixel tightening circuit 604 of FIG.

FIG. 9 is a diagram showing a configuration of a vertical BPF.

FIG. 10 is an edge control circuit 10 according to a fourth embodiment of the present invention.
It is a figure which shows the structure of 1.

FIG. 11 is an edge control circuit 10 according to a fourth embodiment of the present invention.
It is a figure which shows the structure of 1 vertical LPF1705.

FIG. 12 is a diagram for explaining a frequency band in which the edge control circuit 101 of the first and second embodiments of the present invention operates.

FIG. 13 is an edge control circuit 10 according to a third embodiment of the present invention.
1 is a diagram for explaining a frequency band in which 1 operates. FIG.

FIG. 14 is an edge control circuit 10 according to a second embodiment of the present invention.
2 is a characteristic diagram of the horizontal BPF 202, 601 of FIG.

FIG. 15 is an edge control circuit 10 according to a second embodiment of the present invention.
3 is a waveform chart for explaining the operation of No. 1 in FIG.

FIG. 16 is an edge control circuit 10 according to a third embodiment of the present invention.
It is a figure which shows the structure of 1.

FIG. 17 is an edge control circuit 10 according to a third embodiment of the present invention.
FIG. 3 is a waveform diagram for explaining the operation of No. 1.

FIG. 18 is an edge control circuit 10 according to a third embodiment of the present invention.
It is a figure which shows the structure of the normalization circuit 2303 of 1.

FIG. 19 is a diagram showing a configuration of an edge control circuit 101 obtained by combining the third embodiment and the second embodiment of the present invention.

FIG. 20 is a diagram showing a configuration of an edge circuit 101 in which the second embodiment and the third embodiment of the present invention are combined.

FIG. 21 is an edge control circuit 10 according to a fourth embodiment of the present invention.
1 is a diagram for explaining a frequency band in which 1 operates. FIG.

FIG. 22 is a characteristic diagram for explaining the operation of the fourth embodiment of the present invention.

FIG. 23 is a configuration diagram of an application example in which the second embodiment and the fourth embodiment of the present invention are combined.

FIG. 24 is a motion adaptive three-dimensional Y / s according to the fifth embodiment of the present invention.
It is a figure which shows the structure of a C separation circuit.

FIG. 25 is a code determination circuit 2302 and a comparator 20 of FIG.
It is a figure which shows the relationship which combined the output of FIG.

FIG. 26 is an edge control circuit 10 according to a third embodiment of the present invention.
FIG. 6 is a diagram for explaining the operation of the normalization circuit 2303 of No. 1;

FIG. 27 is an edge control circuit 10 according to a third embodiment of the present invention.
FIG. 10 is a diagram for explaining the operation of the vertical BPF 2304 (FIG. 9) of No. 1;

FIG. 28 is an edge control circuit 10 according to a fourth embodiment of the present invention.
It is a figure for demonstrating operation | movement of the vertical LPF175 of 1 (FIG. 11).

FIG. 29 is a motion adaptive three-dimensional Y / s according to the sixth embodiment of the present invention.
It is a figure which shows the structure of a C separation circuit.

FIG. 30 is a color saturation detection circuit 6 according to a sixth embodiment of the present invention.
It is a figure which shows the structure of 000.

FIG. 31 is a color saturation detection circuit 6 according to a sixth embodiment of the present invention.
It is a figure which shows the output relationship of each component of 000.

FIG. 32 is an edge control circuit 60 according to a sixth embodiment of the present invention.
It is a figure which shows the structure of 01.

FIG. 33 is a motion adaptive three-dimensional Y / s according to the seventh embodiment of the present invention.
It is a figure which shows the structure of a C separation circuit.

FIG. 34 is a diagram for specifically explaining the disturbance that occurs when the horizontal edge of the C signal moves in the sixth embodiment of the present invention.

FIG. 35 is a color saturation detection circuit 7 according to a seventh embodiment of the present invention.
FIG. 5 is a diagram for specifically explaining hue detection using the I signal and the Q signal of 000.

FIG. 36 is a color saturation detection circuit 7 according to a seventh embodiment of the present invention.
It is a figure which shows the structure of 000.

FIG. 37 is a color saturation detection circuit 7 according to a seventh embodiment of the present invention.
000 is a diagram showing the configuration of the hue detection circuits 4003 and 4012.

FIG. 38 is a color saturation detection circuit 7 according to a seventh embodiment of the present invention.
4 is a diagram showing a configuration of continuity detection circuits 4004 and 4013 that form the “000”.

FIG. 39 is a motion adaptive three-dimensional Y / s according to the eighth embodiment of the present invention.
It is a figure which shows the structure of a C separation circuit.

FIG. 40 is a diagram for specifically explaining a problem in the ninth embodiment of the present invention.

FIG. 41 is a diagram showing configurations of a horizontal edge detection circuit 8000, a vertical edge detection circuit 8001, and a pixel number detection circuit 8002 according to an eighth embodiment of the present invention.

FIG. 42 is a motion adaptive three-dimensional Y / s according to the ninth embodiment of the present invention.
It is a figure which shows the structure of a C separation circuit.

FIG. 43 is an edge control circuit 90 of the ninth embodiment of the present invention.
It is a figure which shows the structure of 00.

FIG. 44 is a diagram specifically illustrating a case where cross color is likely to occur in the seventh embodiment of the present invention.

FIG. 45 is a diagram showing a configuration of a conventional motion adaptive three-dimensional Y / C separation circuit.

[FIG. 46] An edge detection circuit 1012 and a motion detection circuit 10 between two frames of a conventional motion adaptive three-dimensional Y / C separation circuit
It is a figure which shows the structure of 11.

47 is a characteristic diagram of the non-linear processing circuits 1106 and 1107 of the inter-frame motion detection circuit 1011. FIG.

FIG. 48 is a configuration diagram showing a specific example of a two-dimensional Y / C separation circuit 1005.

[Fig. 49] Fig. 49 is a spectrum diagram showing a frequency region of a luminance signal / color signal of a television signal.

FIG. 50 is a characteristic diagram showing characteristics of 1-frame sum, 1-frame difference, and 2-frame difference.

[Fig. 51] Fig. 51 is a diagram for describing a specific example of interference generated in a conventional motion adaptive three-dimensional Y / C separation circuit.

[Explanation of symbols]

101 ... Edge control circuit, 102, 1013, 70
6 ... MAX circuit, 1001, 1009 ... 525
H delay device, 1002, 1803 ... Adder, 100
3, 1010, 903, 1303, 1306 ... Subtractors, 1004, 202, 601, 1101, 1305,
1702, 2301 ... Horizontal BPF (Band Pa)
ss Filter), 1005 ... Two-dimensional Y / C separation circuit, 1006 ... Horizontal LPF (Low Pass)
Filter), 1007 ... Motion detection circuit between frames 1008, 1012 ... Edge detection circuit, 1
011 ... Motion detection circuit between 2 frames, 1014, 1
015 ... mixer, 1018, 1019, 1304
..Coefficient units, 203, 602, 1102, 1105, 1
703, 1706 ... Absolute value circuit, 204, 603,
1103, 1704, 1707, 2305 ... Comparator, 205, 1708 ... Pixel enlargement circuit, 20
7, 1108 ... Selector, 604 ... Pixel compression circuit, 804 ... MIN circuit, 902, 1802, 13
02 ... 1H delay device, 1106, 1107 ... Non-linear processing circuit, 1705 ... Vertical LPF, 2302 ...
-Code determination circuit, 2303 ... Normalization circuit, 230
4, 3101 ... Vertical BPF, 6000 ... Color saturation detection circuit, 6001 ... Edge control circuit, 3702
... Absolute value circuit, 3703 ... Comparator, 3704,
3705, 3706, 3707 ... 2ck delay device, 3
708 ... AND circuit, 3709 ... 1ck delay device, 3710 ... OR circuit, 3901 ... Selector,
7000 ... Color saturation detection circuit, 4001 ... Color demodulation circuit, 4002 ... Absolute value circuit, 4003 ... Hue detection circuit, 4004 ... Continuity detection circuit, 4005
... Delay circuit, 4006 ... OR circuit, 4007 ...
.... AND circuits, 4008 ... OR circuits, 4009 ...
.... OR circuit, 4010 ... AND circuit, 4011 ...
..Absolute value circuit, 4012 ... Hue detection circuit, 401
3 ... Continuity detection circuit, 4014 ... Delay circuit, 4
01 ... Subtractor, 402 ... Delay device, 403 ...
Absolute value circuit, 404 ... Comparator, 405, 406, 4
07, 408 ... 1ck delay device, 409 ... AND
Circuits, 412 ... Comparators, 413, 414, 415,
416 ... 1ck delay device, 417 ... AND circuit,
8000 ... Horizontal edge detection circuit, 8001 ... Vertical edge detection circuit, 8002 ... Pixel number detection circuit, 8
003 ... AND circuit, 4301 ... Vertical LPF,
4302 ... Horizontal BPF, 4303 ... Absolute value circuit, 4304 ... Comparator, 4305 ... NOR circuit, 4306, 4307, 4308, 4309, 431
0, 4311, 4312, 4313 ... 1ck delay device, 4314 ... Adder, 4315 ... Comparator, 4
317 ... vertical BPF, 4318 ... horizontal LPF,
4319 ... Absolute value circuit, 4320 ... Comparator, 9
000 ... Edge control circuit, 902 ... Horizontal BP
F, 903 ... Absolute value circuit, 904 ... Comparator, 9
05 ... Pixel enlargement circuit, 906 ... AND circuit.

─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Yasuo Takahashi 8 Shinsugita-cho, Isogo-ku, Yokohama-shi, Kanagawa, Toshiba Multimedia Technology Research Laboratory (56) Reference JP-A-5-30530 (JP, A) Flat 3-119886 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 9/44-9/78

Claims (9)

(57) [Claims] 1. A motion adaptive three-dimensional Y for separating a composite video signal into a luminance signal and a chrominance signal according to the amount of motion of an image.
/ C separation circuit, first color signal extraction means for extracting a first color signal from the composite video signal by processing within one field, and second color signal from the composite video signal by processing during one frame Second color signal extraction means for extracting a first motion detection signal for detecting a motion amount of an image from a difference component between frames of the composite video signal, and extracting a first motion detection signal, Second motion detecting means for detecting a motion amount of an image from a difference component between two frames of a composite video signal and extracting a second motion detection signal; and the first color signal and the second color signal. Mixing means for mixing; one-frame sum extraction means for extracting a one-frame sum signal from the composite video signal; and horizontal edge components from the one-frame sum signal. An edge detection unit that generates an edge detection signal indicating the presence or absence of an edge by detecting and detecting whether or not it is larger than a first predetermined value, and the second motion detection signal and a second predetermined value are input. Selecting means for selecting the second predetermined value when the edge detection signal is larger than the first predetermined value, and selecting the second motion detection signal for other cases. An output of the selection means and a first maximum value means for outputting a signal of a level having a larger inner motion amount of the first motion detection signal, wherein the mixing operation of the mixing means is the first maximum value. Motion adaptive three-dimensional Y characterized by being controlled by the output of the value means
/ C separation circuit. The method according to claim 2, wherein said edge detection means comprises a first filter means for extracting a horizontal high-frequency component, the first fill
The data means corresponds to the horizontal high frequency component when the color signal is demodulated.
2. The maximum gain is provided in a certain band.
3. A motion adaptive three-dimensional Y / C separation circuit described in 1. 3. The edge detector is configured to detect horizontal edges.
The component is the first place The above-mentioned d
Second maximum value that expands the horizontal area of the edge detection signal
The device according to claim 1 or 2, further comprising a means.
On-board motion adaptive 3D Y / C separation circuit. 4. The edge detection means is the second maximum
The edge detection signal expanded in the horizontal direction by the value means.
Optimize the area of the No. by shrinking horizontally
4. A third optimization means is provided.
3. A motion adaptive three-dimensional Y / C separation circuit described in 1. 5. The amount of motion of the composite video signal
Motion-adaptive three-dimensional Y that separates into luminance signal and chrominance signal according to
In the / C separation circuit, The composite video signal is processed by processing within one field.
Color signal extraction means for extracting a first color signal from the signal
When, The composite video signal is processed by processing for one frame.
Second color signal extraction means for extracting a second color signal from Difference component between one frame of the composite video signal
Detects the amount of motion of the video and extracts the first motion detection signal
First motion detecting means for Difference component between two frames of the composite video signal
Detects the amount of motion in the image and extracts the second motion detection signal
Second motion detecting means for Mixer for mixing the first color signal and the second color signal
Dan, Sum signal for one frame from the composite video signal
1-frame sum extraction means for extracting Difference signal between one frame from the composite video signal
1 frame difference extraction means for extracting Detect a horizontal edge component from the one-frame sum signal
The first edge signal and the one-frame difference signal
The second edge that detects the vertical low-frequency component of the flat edge component
Generates an edge detection signal that indicates the presence or absence of an edge from the signal
Edge detection means for The second motion detection signal and the first predetermined value are input,
The edge detection signal is the first edge signal at the second position.
If the second edge signal is larger than a predetermined value and the third predetermined value is
When the value is lower, select the first predetermined value and
Edge detection signal If the number indicates otherwise, the second action
Selection means for selecting a detection signal, Output of the selecting means and internal motion of the first motion detection signal
With the maximum value means that outputs the signal of the level with the larger amount
Equipped with, The mixing operation of the mixing means depends on the output of the maximum value means.
Motion-adaptive three-dimensional Y / C component characterized by being controlled
Isolated circuit. 6. The amount of motion of the composite video signal
Motion-adaptive three-dimensional Y that separates into luminance signal and chrominance signal according to
In the / C separation circuit, The composite video signal is processed by processing within one field.
Color signal extraction means for extracting a first color signal from the signal
When, The composite video signal is processed by processing for one frame.
Second color signal extraction means for extracting a second color signal from Difference component between one frame of the composite video signal
Detects the amount of motion of the video and extracts the first motion detection signal
First motion detecting means for Difference component between two frames of the composite video signal
Detects the amount of motion in the image and extracts the second motion detection signal
Second motion detecting means for Mixer for mixing the first color signal and the second color signal
Dan, Sum and division for one frame from the composite video signal
1-frame sum extraction means for extracting the number, Detect horizontal edge component from the one-frame summation signal
However, by detecting whether or not it is larger than the first predetermined value,
Edges that generate edge detection signals that indicate the presence or absence of
Detection means, A first controlling the first predetermined value from the first color signal;
Control means of The second motion detection signal and the second predetermined value are input,
Indicates that the edge detection signal is larger than the first predetermined value.
When the second predetermined value is selected and other values are indicated
Means for selecting the second motion detection signal; The amount of movement of the output of the selecting means and the first movement signal
The maximum value means for outputting the signal of the larger level of
Be equipped with The mixing operation of the mixing means depends on the output of the maximum value means.
Motion-adaptive three-dimensional Y / C component characterized by being controlled
Isolated circuit. 7. The first control means controls the first color signal.
Color saturation detection means for detecting the saturation level of the
Is high, the first predetermined value is high, and the saturation is low.
Second control means for controlling the first predetermined value to be low
7. The movement suiting device according to claim 6, further comprising a step.
Three-dimensional Y / C separation circuit. 8. The amount of motion of the composite video signal
Motion-adaptive three-dimensional Y that is separated into a luminance signal and a color signal according to
In the / C separation circuit, The composite video signal is processed by processing within one field.
Color signal extraction means for extracting a first color signal from the signal
When, The composite video signal is processed by processing for one frame.
Second color signal extraction means for extracting a second color signal from Difference component between one frame of the composite video signal
Detects the amount of motion of the video and extracts the first motion detection signal
First motion detecting means for Difference component between two frames of the composite video signal
Detects the amount of motion in the image and extracts the second motion detection signal
Second motion detecting means for Mixer for mixing the first color signal and the second color signal
Dan, Sum and division for one frame from the composite video signal
1-frame sum extraction means for extracting the number, Detect horizontal edge component from the one-frame summation signal
However, by detecting whether or not it is larger than the first predetermined value,
Edges that generate edge detection signals that indicate the presence or absence of
Detection means, The first color signal and the composite video signal
And a control means for controlling the first predetermined value, The second motion detection signal and the second predetermined value are input,
Indicates that the edge detection signal is larger than the first predetermined value.
When the second predetermined value is selected and other values are indicated
Means for selecting the second motion detection signal; The amount of movement of the output of the selecting means and the first movement signal
The maximum value means for outputting the signal of the larger level of
Be equipped with The mixing operation of the mixing means depends on the output of the maximum value means.
Motion-adaptive three-dimensional Y / C component characterized by being controlled
Isolated circuit. 9. The control means controls the saturation of the first color signal.
It has a high degree of harmony, and is it the vertical low range of the video composite signal?
The level of the absolute value of the horizontal high frequency component and the vertical high frequency and horizontal
If the level of the absolute value of the low frequency component is low, the first
The set value is set to a high value, and the set value is set according to claim 8.
Motion adaptive 3D Y / C separation circuit.