JPH07154820A - Luminance signal/chrominance signal separating device between frames - Google Patents

Abstract

PURPOSE:To reduce memory capacity by half as compared with conventional one, while securing separation performance. CONSTITUTION:A sub-sampling is performed for a composite video signal by a sub-sampling circuit 2 and a delay by a one-frame period is performed for the signal in a delay circuit 3. The input/output signal of the delay circuit 3 is inputted in a subtraction circuit 4 and a chrominance signal is calculated. Each chrominance signal of the signal phases omitted by the sub-sampling is interpolated and calculated by a horizontal interpolation circuit 5 and a vertical interpolation circuit 6 based on the output signal of the subtraction circuit 4. The output signals of the subtraction circuit 4, the horizontal interpolation circuit 5 and the vertical interpolation circuit 6 are adaptively selected in a selection circuit 8 by the timing pulse synchronized with the timing of the sub-sampling of the sub-sampling circuit 2 and the control by the output result of a correlation detection circuit 7, and the signals are outputted as separated chrominance signals from an output terminal 9.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for separating a composite video signal into a luminance signal and a carrier color signal.

[0002]

2. Description of the Related Art A luminance signal / color signal for separating a composite video signal in which a luminance signal (hereinafter referred to as Y) and a color signal (hereinafter referred to as C) are frequency-multiplexed into an original luminance signal and a color signal. Separation processing (hereinafter referred to as Y / C separation) is conventionally called processing in the horizontal scanning direction only, and is therefore called one-dimensional processing.
It was performed using a low pass filter and a high pass filter. In the case of this configuration, complete separation cannot be performed, and there is a problem of image quality deterioration due to occurrence of cross color and dot interference.

Therefore, since the processing is performed in the horizontal scanning direction and the vertical scanning direction, so-called two-dimensional processing is improved by the processing using the inter-line comb filter that utilizes the inter-line correlation of the composite video signal. Improvements have been made by means of an adaptive process in which the comb filter and the one-dimensional filter of 1 are used to adaptively select them.

However, even by such processing, there is a problem that the resolution in the diagonal direction is lowered due to the limitation of the frequency characteristic due to the calculation in the horizontal and vertical directions,
Complete separation could not be performed.

In order to improve such problems, in recent years,
There is a so-called three-dimensional processing method in which processing is performed in the time axis direction using inter-frame correlation of a composite video signal. With this method, a still image can be completely separated into a luminance signal and a chrominance signal.

FIG. 10 is a schematic view of an NTSC composite video signal as viewed in the direction perpendicular to the time axis. In FIG. 10, vertical lines indicate fields, and circles on the vertical lines indicate lines. Further, Y means a luminance signal and C means a color signal. Y + C means a composite video signal in which the phase of the color signal is positive, and Y-C means a composite video signal in which the phase of the color signal is negative.

As shown in FIG. 10, the NTSC composite video signal is characterized in that the phase of the color signal is inverted between lines and frames. Therefore, when the luminance signal is Y and the chrominance signal is C, the relationship of the signal contents between the point a of interest and the point b separated by one frame period from that point is Ya + C
It becomes a relation of a and Yb-Cb.

If the image is a still image, then Ya
And Yb are equal, and Ca and Cb are equal,
It can be seen as a relationship between Y + C and Y-C.

In this case, a complete color signal can be separated by subtracting the signal at the point b from the signal at the point a, attenuating the amplitude thereof to ½, and returning to the original amplitude.

The above processing is expressed by an equation (Equation 1).

[0011]

[Equation 1]

Then, the Y signal can be obtained by subtracting the C signal thus obtained from the original Y + C signal.

According to this method, when the image is a still image, a complete separation process can be performed without degrading the image quality. This method is called inter-frame Y / C separation processing and has the configuration shown in FIG.

In FIG. 11, 21 is an input terminal,
A composite video signal is provided. Reference numeral 22 denotes a frame memory, which supplies the composite video signal input from the input terminal 21 to the subtractor 23 after delaying by one frame period. The subtractor 23 subtracts the composite video signal delayed by one frame period by the frame memory 22 from the composite video signal input to the input terminal 21, and supplies the subtraction result to the multiplier 24. The multiplier 24 multiplies the signal input from the subtractor 23 by 1/2 its amplitude, calculates the color signal returned to the original amplitude, and outputs the result as the color signal separated from the output terminal 25. It outputs and supplies it to the subtractor 26. The subtractor 26 subtracts the signal from the multiplier 24 from the signal supplied from the input terminal 21 via the delay circuit 27 to calculate the luminance signal. This luminance signal is output from the output terminal 28 as a separated luminance signal.

The delay circuit 27 is inserted in order to match the time axes of the color signal from the multiplier 24 and the composite video signal from the input terminal 21.

As described above, the Y / C separation between frames can completely separate a still image, but a frame memory is required to realize it. This frame memory has a capacity of 4 Mbits in the case of a signal sampled at 4 fsc, and has a large capacity. Therefore, there are problems that the circuit scale becomes large and the cost increases.

[0017]

As described above, the luminance signal / chrominance signal separation utilizing the inter-frame correlation of the composite video signal can achieve complete separation for a still image.
For that purpose, there is a problem that the circuit scale becomes large because a memory capacity for holding at least a signal for one frame is required.

Therefore, the present invention is to provide an interframe luminance signal / color signal separation device capable of reducing the memory capacity by half while having the same performance as the conventional interframe luminance signal / color signal separation performance. To aim.

[0019]

In order to achieve the above object, the inter-frame luminance signal / chrominance signal separation device of the present invention is a composite sampled at a frequency 4n times (n is a positive integer) the color subcarrier frequency. Subsampling means for subsampling 2n pixels consecutively every 2n pixels in the horizontal direction with respect to the video signal in accordance with the pixel period and the scanning line period, and a first delay for delaying one frame period using the subsampling result as an input. Means for calculating the color signal component by calculating the values of the composite video signal and the output result of the delay means, and the signal phase missing by the sub-sampling means using the color signal component as input. Horizontal interpolation means for interpolating and calculating color signal components in the horizontal direction, and missing by the sub-sampling means using the color signal components as input The vertical interpolating means for interpolating and calculating the color signal component in the signal phase in the vertical direction has a stronger correlation in the horizontal direction or the vertical direction in the signal phase missing by the sub-sampling means. Correlation detection means for detecting whether the color signal component, the horizontal interpolation color signal component and the vertical interpolation color signal component as input, the presence or absence of a signal missing in the sub-sampling means and the output result of the correlation detection means Selecting means for selecting and outputting one of the three input signals to obtain a color separation signal, and second delay means for delaying the composite video signal by a specific period.
Subtraction means for subtracting the color separation signal from the output signal of the second delay means to obtain a luminance separation signal.

[0020]

According to the present invention, the number of data of the composite video signal is reduced by sub-sampling by the above configuration, and the missing of the signal caused by the sub-sampling is compensated by the interpolation processing in the horizontal direction and the interpolation processing in the vertical direction. As a result, it is possible to reduce the used memory capacity while ensuring the same level of inter-frame luminance signal / color signal separation performance as in the past.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a luminance signal / color signal separating apparatus according to an embodiment of the present invention.

In FIG. 1, reference numeral 1 is an input terminal to which an NTSC composite video signal sampled at a frequency four times that of a color subcarrier is supplied.

This composite video signal is input to the subsampling circuit 2 and subsampled in accordance with the cycle of the color subcarrier of the input composite video signal. Reference numeral 3 denotes a first delay circuit, which writes to a sub-sampled signal input only when a signal is present, holds this for one frame period, and outputs after one frame period has elapsed. The output of the delay circuit 3 is input to one of the subtraction circuits 4. The composite video signal input to the input terminal 1 is input to the other input of the subtraction circuit 4. The subtraction circuit 4 calculates a color signal by subtracting the output signal of the delay circuit 3 from the composite video signal input to the input terminal 1.

The horizontal interpolation circuit 5 and the vertical interpolation circuit 6 receive the output signal of the subtraction circuit 4 as input, and interpolate and calculate the color signals at the signal phase missing due to subsampling.

The correlation detection circuit 7 receives the composite video signal input from the input terminal 1 and the color signal output from the subtraction circuit 4 as inputs, and the horizontal and vertical directions at the signal phase missing in the sub-sampling circuit 2. Detect which is more correlated.

Reference numeral 8 denotes a selection circuit, which receives the output signal of the subtraction circuit 4 and the output signals of the horizontal interpolation circuit 5 and the vertical sampling circuit 6 and correlates with a timing pulse synchronized with the subsampling timing of the subsampling circuit 2. One of the input signals is output under the control of the output result of the detection circuit 7. In this way, the output of the selection circuit 8 becomes a color signal with no omission, and is output from the output terminal 9 as a separated color signal and is input to the subtraction circuit 11.
The composite video signal from the input terminal 1 is input to the other input of the subtraction circuit 11 via the second delay circuit 10. The subtraction circuit 11 calculates the luminance signal by subtracting the color signal input from the selection circuit 8 from the composite video signal input from the second delay circuit 10, and the output terminal 1
It is output from 2 as a separated luminance signal.

The second delay circuit 10 includes a selection circuit 8
Is inserted in order to match the time axes of the color signal from the input terminal 1 and the composite video signal from the input terminal 1.

The overall construction of one embodiment has been described above, and its operation will be described below.

FIG. 2 is a schematic diagram showing signal components of pixels in a field when sampling is performed at a frequency four times as high as a color subcarrier, and FIG. 3 is a schematic diagram showing an example of a signal form in the subsampling processing of the present invention. It is a figure.

For example, if the sampling phase is synchronized with the phases of the I and Q axes of the color signal in the composite video signal, the signal contents are as shown in FIG. 2, ..., Y + I, Y + Q,
A signal in which I-axis components and Q-axis components such as Y-I, Y-Q, ... Are alternately time-axis multiplexed.

In the present embodiment, the sub-sampling circuit 2 responds to such signals by, for example, ..., Y + I, Y.
A sub-sampling output including a positive I-axis component and a Q-axis component such as + Q, none, none, Y + I, Y + Q, ... Is obtained.

As shown in FIG. 3, this sub-sampling has a grid pattern in which two pixels which are continuous in the horizontal direction of the input signal are viewed as a group and are alternately present and absent in the horizontal and vertical directions. It will be.

The output result of the sub-sampling circuit 2 is input to the first delay circuit 3, is delayed by one frame period, and is output to the subtraction circuit 4. Regarding the delay circuit 3, the input signal has only half of the original number of pixels. Therefore, if the input is limited only to the presence of pixels, the capacitance of the delay circuit 3 can be set to one field, which is one half of the original capacity.

The subtraction circuit 4 subtracts the input signal from the delay circuit 3 from the composite video signal from the input terminal 1. Since the two input signals of the subtraction circuit 4 are separated by one frame period, the subtraction result is an inter-frame color separation result utilizing the fact that the phase of the color signal is inverted between frames in the composite video signal. Become.

However, the color signal separated and processed between the frames exists only in the signal phase selected by the subsampling. In other words, regarding the signal phase missing by the sub-sampling circuit 2, there is no signal necessary for color signal separation processing between frames of the composite video signal, so the color signal obtained by the above calculation is the color signal of the sub-sampling output. This is only during the signal phase, and the missing color signal phase exists in the color signal that has been subjected to the separation processing.

Explaining the existence state of this color signal, in the sequence of signal intervals of four times the frequency of the color subcarrier (hereinafter referred to as 4fsc), when existence is indicated by ◯ and non-existence is indicated by x, the horizontal direction of the screen Then ... ○○ ×× ○○ ×× ... becomes a sequence, and in the field vertical direction within the field ... ○ × ○ ×
... will be.

Further, when the color signal is regarded as a combination of the I-axis component and the Q-axis component, the color signal component is as shown in FIG.

Therefore, the components of the color signal are originally ..., I, Q, -I, -Q, I, Q,-in the horizontal direction of the screen.
For I, -Q, ..., I, Q,
X, x, I, Q, x, x, ... exist, and in the vertical direction of the screen in the field, originally ..., I,-
I, I, -I, ... Or ..., Q, -Q, Q,-
For Q, ..., I, ×, I,
X, ... Or ..., Q, x, Q, x, ... Or by subsampling phase ..., x,-
I, ×, −I, ... Or ..., ×, −Q, ×, −
It will exist like Q, ...

That is, paying attention to each color axis component, what originally has positive polarity and negative polarity signals alternately becomes a signal sequence having only positive polarity or negative polarity by this sub-sampling.

In consideration of this relationship, the horizontal interpolation circuit 5 and the vertical interpolation circuit 6 interpolate and calculate the color signal components at the signal missing points of the color signal calculation result having this missing as follows.

First, the horizontal interpolation circuit 5 calculates an average value between two pixels which have the same color axis component as the missing point and are separated from the missing point by two pixels to the left and right respectively, and the sign of the average value is inverted. Get the interpolation result. The positional relationship of the signals in this case is shown in FIG. 4, and the pass frequency band of the signals to be interpolated is shown in FIG.

The vertical interpolation circuit 6 calculates an average value between two pixels having the same horizontal position one line above and one line below the missing point in the field having the same color axis component as the missing point. The result of interpolation is obtained by inverting the sign of. The positional relationship of the signals in this case is shown in FIG. 6, and the pass frequency band of the interpolated signal is shown in FIG.

The above interpolation processing has the following advantages and disadvantages in the performance of the processing result. This is because in the horizontal interpolation process, dot interference occurs in the case of a pattern having a non-correlation in the horizontal direction such as the color boundary part in the horizontal direction of the color bar, and in the vertical interpolation process, the color boundary in the vertical direction is generated. In the case of a pattern having vertical non-correlation such as a part, dot interference occurs.

This is due to the deterioration of the frequency characteristics due to the pass band limitation of the interpolation processing in the interpolation direction.

For example, in the case of a pattern having a decorrelation in the horizontal direction and a correlation in the vertical direction, such as the color boundary portion in the horizontal direction of the color bar, the frequency component of the color signal is as shown in FIG.
As described above, the horizontal frequency is large and the vertical frequency is small.

In such a picture state, if interpolation processing is performed in the vertical direction as in the vertical interpolation circuit 6 of this embodiment, the color signal components can be calculated without omission, so that dot interference does not occur. Absent.

Further, in the case of a pattern having a correlation in the horizontal direction and a non-correlation in the vertical direction such as a color boundary portion in the vertical direction, the frequency component of the color signal has a horizontal frequency as shown in FIG. It becomes small on the side and widens on the vertical frequency side.

In such a picture state, if interpolation processing in the horizontal direction as in the horizontal interpolation circuit 5 of this embodiment is performed, the color signal components can be calculated without omission, so that dot interference does not occur. Absent.

Therefore, according to the present invention, the strength of correlation in the horizontal direction and the strength in the vertical direction are detected, and the interpolation processing result in the direction having a stronger correlation is selected, thereby performing the interpolation processing in each of the horizontal and vertical interpolations. Complement each other's performance problems and utilize only the advantages.

As a method of detecting this correlation, in the present embodiment, a difference value between data used for horizontal interpolation and a difference value between data used for vertical interpolation are respectively calculated. Then, by comparing the two, it is determined that there is a stronger correlation in the direction of the smaller difference value, and the magnitude of each difference value indicates that the larger the difference value, the stronger the decorrelation. Both the judgment of the decorrelation and the judgment of the correlation are used to determine the overall correlation. Then, based on this comprehensive correlation, the interpolation processing in the direction determined to have a stronger correlation is selected, and the optimal processing is adaptively performed for the pattern.

As described above, in this embodiment, the composite video signal is sub-sampled and supplied to the delay circuit, and the sub-sampling is performed in the color signal output by the inter-frame processing between the input and output signals of the delay circuit. The signal missing portion due to is interpolated with a nearby signal, and the interpolation is performed by an adaptive process using a pattern in consideration of frequency characteristics in the horizontal and vertical directions.

With such a configuration, the capacitance of the first delay circuit 3 has the number of pixels halved by sub-sampling.
Since it suffices to store the signal reduced to, it can be reduced to one half of the conventional one.

Further, in this embodiment, since the signal dropout due to the subsampling is compensated by the interpolation processing of the horizontal catching circuit 5 and the vertical catching circuit 6, the number of pixels is reduced by the subsampling. It is possible to obtain a separation performance that is almost the same as the conventional one.

Other than this embodiment, various modifications can be made without departing from the scope of the present invention.

[0056]

As described above, according to the present invention, while the performance equivalent to that of the device using the conventional memory capacity for one frame is maintained, the required memory capacity is 2
It is possible to realize an inter-frame luminance signal / color signal separation device that is one-third.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a luminance signal / color signal separation device according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing signal components of pixels in a field when an NTSC composite video signal is sampled at 4 fsc.

FIG. 3 is a schematic diagram showing an example of a signal form in subsampling processing of the present invention.

FIG. 4 is a schematic diagram showing the positional relationship of signals in the horizontal interpolation processing of the present invention.

FIG. 5 is a characteristic diagram showing a pass band in the two-dimensional frequency domain of the horizontal interpolation processing of the present invention.

FIG. 6 is a schematic diagram showing the positional relationship of signals in the vertical interpolation processing of the present invention.

FIG. 7 is a characteristic diagram showing a pass band in the two-dimensional frequency domain of the vertical interpolation processing of the present invention.

FIG. 8 is a characteristic diagram showing a band in a two-dimensional frequency region of a pattern having a decorrelation in the horizontal direction.

FIG. 9 is a characteristic diagram showing a band in a two-dimensional frequency region of a pattern having decorrelation in the vertical direction.

FIG. 10 is a schematic diagram showing characteristics of a composite video signal.

FIG. 11 is a block diagram showing the configuration of a conventional inter-frame luminance signal / color signal separation device.

[Explanation of symbols]

 1 Input Terminal 2 Sub Sampling Circuit 3,10 Delay Circuit 4,11 Subtraction Circuit 5 Horizontal Interpolation Circuit 6 Vertical Interpolation Circuit 7 Correlation Detection Circuit 8 Selection Circuit 9 Color Signal Output Terminal 12 Luminance Signal Output Terminal

Claims (1)

[Claims] 1. A composite video signal sampled at a frequency of 4n times the color subcarrier frequency (n is a positive integer) is divided into 2n pixels continuous every 2n pixels in the horizontal direction according to the pixel period and the scanning line period. A sub-sampling means for sampling, a first delay means for receiving the sub-sampling result as an input and delaying it by one frame period, and a color signal for calculating the values of the composite video signal and the output result of the delay means as inputs. A calculation means for obtaining a component; a horizontal interpolation means for inputting the color signal component to interpolate a color signal component in the signal phase missing by the sub-sampling means in the horizontal direction; and a sub-sampling using the color signal component as an input. Vertical interpolating means for interpolating and calculating in the vertical direction the color signal component in the signal phase missing by the means; Correlation detection means for detecting which of the correlation in the horizontal direction and the correlation in the vertical direction in the signal phase missing due to the step has a stronger correlation, the color signal component, the horizontal interpolation color signal component, and Selecting means for selecting and outputting one of the three input signals according to the presence or absence of a signal missing in the sub-sampling means and the output result of the correlation detecting means using the vertically interpolated color signal component as an input; A second delay unit that delays the composite video signal by a specific period; and a subtraction unit that subtracts the color separation signal from the output signal of the second delay unit to obtain a luminance separation signal. Inter-frame luminance signal / color signal separation device.