JPH0646814B2 - NTSC adaptive contour extraction filter - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an NTSC adaptive contour extraction filter for digitally extracting a contour signal from an NTSC composite video signal in which a luminance signal and a chrominance signal are frequency-multiplexed. .

[Prior Art] Various methods have been proposed in various fields for extracting the contour component from image information. For example, in a television receiver, a contour signal is extracted from a luminance signal and the original luminance is extracted. The sharpness of the image is improved by adding it to the signal.

The composite video signal S (t) of the NTSC system is the luminance signal Y (t),
Color subcarrier fsc = 3 for the two color difference signals U (t) and V (t).
It is a composite signal with the color signal C (t) that is quadrature-phase modulated at 579545 MHz. That is, S (t) = Y (t) + C (t) = Y (t) + U (t) Sin2πfsct + V
(t) cos2πfsct A conventional contour extraction filter of this type, which may be of analog type or digital type, separates the above-mentioned luminance signal Y (t) from the above-mentioned composite video signal S (t). However, with respect to the separated luminance signal Y (t), a horizontal contour signal is obtained by a horizontal contour extraction filter, and a vertical contour signal is obtained by a vertical contour extraction filter. The horizontal contour extraction filter and the vertical contour extraction filter will be described below.

FIG. 4 is a diagram showing an example of a conventional digital horizontal contour extraction filter. In FIG. 4, an analog composite video signal is applied from the input terminal 1 to the A / D converter 2. The A / D converter 2 converts an analog signal into a digital signal, and a luminance signal color signal separation circuit 3 that separates the converted digital signal into a luminance signal and a color signal.
(Hereinafter referred to as YC separation circuit). YC separation circuit 3
The luminance signal separated by is the horizontal contour extraction filter 8
Given to. The horizontal contour extraction filter 8 is composed of a first delay circuit 51, a second delay circuit 52, a coefficient circuit 6 and an addition circuit 7. The contour signal extracted from the horizontal contour extraction filter 8 is output to the output terminal 4.

Next, the operation of the horizontal contour extraction filter shown in FIG. 4 will be described.

The analog composite video signal input to the input terminal 1 is A /
The signal is supplied to the D converter 2 and is converted into a digital composite video signal at the predetermined sampling frequency fs. The digital composite video signal output from the A / D converter 2 is separated into a luminance signal and a chrominance signal by the YC separation circuit 3, and the separated luminance signal is the first delay circuit of the horizontal contour extraction filter 8. In addition to being given to 1, it is also given to the adder circuit 7. The output of the first delay circuit 51 is
The coefficient circuit 6 is supplied to the second delay circuit 52.
And the output signal of the first delay circuit 51 is multiplied by "-2" in the coefficient circuit 6. Here, the delay time of each of the first and second delay circuits 51 and 52 is determined by the reciprocal of the sampling frequency fs, and one sampling interval of the digital signal sequence output from the A / D conversion circuit 2 It is set to be T.

In the adder circuit 7, the output of the YC separation circuit 3, the output of the coefficient circuit 6 and the output of the second delay circuit 52 are added. Therefore, assuming that the luminance signal f (t) which is the output signal of the YC separation circuit 3 is f (nT) at a time t = nT, as will be understood from the above description, the output of the addition circuit 7 is , F (nT) -2f {(n-1) T} + f {(n-2)
T} is obtained. This is the second-order differential of the luminance signal f (t) in the horizontal direction on the screen, and therefore the horizontal high-frequency component of the luminance signal, that is, the horizontal contour signal is extracted.

FIG. 5 is a diagram showing an example of a conventional digital vertical contour extraction filter. In FIG. 5, the vertical contour extraction filter 9 is constituted by the third and fourth delay circuits 53 and 54, the coefficient circuit 6 and the addition circuit 7, and the A / D converter 2
The YC separation circuit 3 is similar to that shown in FIG. The delay times of the third and fourth delay circuits 53 and 54 are configured to be one horizontal scanning time, and the coefficient circuit 6 and the addition circuit 7
Is constructed in the same manner as in FIG. 4, the coefficient circuit 6 multiplies the output of the third delay circuit 53 by "-2", and the adder circuit 7 produces YC.
The output of the separation circuit 3, the output of the coefficient circuit 6 and the output of the fourth delay circuit 54 are added. Therefore, as is clear from the description in FIG. 4, the vertical contour extraction filter shown in FIG. 5 performs the second-order differentiation in the vertical direction on the screen with respect to the luminance signal which is the output signal of the YC separation circuit 3. And then
A vertical high frequency component of the luminance signal, that is, a vertical contour signal is extracted.

FIG. 6 is a diagram showing a conventional contour extraction filter constructed by superposing the horizontal contour extraction filter shown in FIG. 4 and the vertical contour extraction filter shown in FIG. In FIG. 6, a contour extraction filter 10 is a filter for extracting contour signals in the horizontal and vertical directions, and the first and second delay circuits 51 and 52 are the first and second delay circuits described in FIG. The delay circuits 51 and 52 have the same delay time.
The fifth and sixth delay circuits 55 and 56 are configured to have a delay time obtained by subtracting the delay times of the first and second delay circuits 51 and 52 from one horizontal scanning time.
Therefore, the total delay time of the delay time of the first delay circuit 51 and the delay time of the fifth delay circuit 55 is one horizontal scanning time, and the delay time of the second delay circuit 52 and the sixth delay circuit 56. The delay time including the delay time of 1 is one horizontal scanning time.

The coefficient circuit 6 is configured to multiply the input signal by "-4", and the adder circuit 7 outputs the output of the YC separation circuit 3, the output of the fifth delay circuit 55, the output of the coefficient circuit 6 and the second delay circuit 52.
Is added to the output of the sixth delay circuit 56. That is, the contour extraction filter 10 uses the second
Vertical contour extraction filter including the delay circuit 55, the first delay circuit 51, the second delay circuit 52, the sixth delay circuit 56, the coefficient circuit 6, and the adder circuit 7, and the first delay circuit 51. , The second delay circuit 52, the coefficient circuit 6, and the horizontal contour extraction filter constituted by the adder circuit 7 are superposed, whereby the vertical contour signal and the horizontal contour signal can be extracted.

[Problems to be solved by the invention]

As described above, since the conventional contour extraction circuit extracts the contour signal using the luminance signal, the YC separation circuit that separates the composite video signal into the luminance signal and the color signal delays one horizontal scanning time signal. When a circuit is used, this delay circuit and the delay circuit for delaying one horizontal scanning time signal necessary for the contour extraction filter cannot be shared, and there is a drawback that an increase in cost cannot be avoided.

The present invention has been made in order to solve the above problems, and by extracting a contour signal directly from a composite video signal, a delay circuit required for a YC separation circuit and a delay circuit required for a contour extraction filter are provided. And an inexpensive contour signal extraction filter that can be shared.

[Means for solving problems]

The adaptive contour extraction filter according to the present invention is located above and below the sampling point of the digital signal obtained by A / D converting the composite video signal of the NTSC system, and the color subcarrier phases are in phase with each other. , And the lower and upper sampling points, which are in antiphase with the sampling point, are located on the left and right of the sampling point on the same horizontal scanning line as the sampling point, and the color subcarrier phases are in phase with each other, and With respect to the left and right sample points of the opposite phase, a delay means for extracting each sample value at the same timing, three sample values of the sample point and the upper and lower sample points are used, First signal change amount detection means for detecting a signal change amount in the vertical direction in which these three sample points are arranged,
A second signal change amount detection means for detecting the signal change amount in the horizontal direction in which these three sample points are arranged by using the sample points and the three sample points of the left and right sample points, One of the vertical and horizontal contour signals is extracted and output based on the magnitude comparison between the vertical signal change amount and the horizontal signal change amount.

[Action]

In the present invention, the sampled value of the sampled point and the sampled values of the upper and lower sample points, which are in phase with each other in the color subcarrier phase and opposite in phase to the sampled point, in the vertical direction from the sampled value of the lower sample point. While detecting the signal change amount, from the sampled value of the sampled point and the sampled values of the left and right sampled points that are located on the left and right of the sampled point and have the same color subcarrier phase and the opposite phase of the sampled point. The amount of signal change in the horizontal direction is detected, and by comparing the amount of signal change in the vertical direction with the amount of signal change in the horizontal direction, either the vertical or horizontal contour signal is extracted and output. Therefore, in the circuit configuration in which both the Y / C separation system and the contour extraction system are integrated, the line delay circuit required for the YC separation circuit and the line delay circuit required for the contour extraction filter can be shared, and the line delay circuit can be reduced. Can

Further, the detection of the image change direction is performed by comparing the magnitudes of the differential output values in the vertical and horizontal directions, that is, the difference between the sample values of the upper and lower sample points and the difference between the sample values of the left and right sample points that are close to the sample point. Not only that, the comparison is performed by comparing the magnitude of the signal change amount including the sample value of the sample point. Therefore, it is possible to feed back the situation of the sample point and more faithfully extract the contour of the sample point.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing an embodiment of the present invention. In the figure, 1 is an input terminal, and a composite video signal is given to the A / D converter 2 via the input terminal 1. This A /
The D converter 2 is supplied with a sampling pulse having a frequency fs which is four times the color subcarrier frequency fsc of the composite video signal from the pulse generator 30, and the A / D converter 2 follows the composite video according to the sampling pulse. The signal is converted into a digital signal, which is applied to the contour extraction filter 11. Contour extraction filter 1
1 is the first and second delay circuits 201 and 202 and the third and third delay circuits 201 and 202.
Fourth delay circuits 21 and 22 and first and second adder circuits 2
21, 222, 1/4 multiplication circuit 23, first to third subtraction circuits 241-243, and first and second absolute value circuit 25.
1, 252, a comparison circuit 26, a sign inversion circuit 27,
It is composed of a multiplication circuit 28 and a doubling circuit 29.

The first and second delay circuits 201 and 202 described above delay the sample value series output from the A / D converter 2 by a time obtained by subtracting two sample periods from one horizontal scanning time. The third and fourth delay circuits 21 and 22 are A / D converters 2.
The sampled value sequence which is the output of is delayed by two sample periods. The first adder circuit 221 is the first delay circuit 2
01 output and the output of the third delay circuit 22 are added, and the second adder circuit 222 adds the output of the A / D converter 2 and the output of the second delay circuit 202. . The first subtraction circuit 241 is a circuit that subtracts the output signal of the second addition circuit 222 from the output signal of the first addition circuit 221. The 1/4 multiplication circuit 23 is configured to multiply the output of the first subtraction circuit 241 by 1/4. The doubling circuit 29 doubles the output of the third delay circuit 21. The second and third subtraction circuits 242 and 243 are configured to subtract the output of the doubling circuit 29 from the outputs of the first and second addition circuits 221 and 222, respectively.

The first absolute value circuit 251 is for obtaining the absolute value of the output of the subtraction circuit 243, and the second absolute value circuit 252.
Is the absolute value of the output of the subtraction circuit 242. The sign inversion circuit 27 outputs 1 when the comparison output of the comparison circuit 26 is reached.
The sign of the output of the / 4 multiplication circuit 23 is inverted.

FIG. 2 is a diagram showing a sampling signal sequence of an NTSC composite video signal sampled by a sampling pulse, and FIG. 3 is a diagram for explaining a specific operation of the embodiment of the invention. Therefore, the sample value series shown in FIG. 2 is marked with a symbol. In addition, ◎ in FIG. 2 and FIG. Indicates that the phases of the color subcarriers are reversed.

Next, the specific operation of the embodiment of the present invention will be described with reference to FIGS. First, the contour extraction of the sample point P3 shown in FIG. 3 will be described. Now, at a certain time T, if the sample value of the sample point P5 is output from the A / D converter 2, the sample value of the sample point P4 is output from the first delay circuit 201, and the sample value of the sample point is output from the second delay circuit 202. The sample value at the point P1 and the sample value at the sample point P2 are output from the fourth delay circuit 22, respectively. Then, in the first adder circuit 221, the output (P4) of the first delay circuit 201 and the fourth delay circuit 2
Second output (P2) is added, and the second addition circuit 222 is added.
Then, the output (P5) of the A / D converter 2 and the second delay circuit 2
02 output (P1) is added, and the first subtraction circuit 24
At 1, the output of the second adder circuit 222 (P1 + P5) is subtracted from the output of the first adder circuit 221 (P2 + P4), and the 1/4 multiplying circuit 23 outputs this first subtractor circuit 2
The output (P2 + P4-P1-P5) of 41 is multiplied by 1/4. Therefore, the output signal of the 1/4 multiplication circuit 23 is -1/4 (sample value of the sample point P1) -1/4 (sample value of the sample point P5) +1/4 (sample value of the sample point P2) +1 / 4 (sample value of sample point P4) (1). It can also be written as

[-1/4 (sample value of sample point P1) + 1/2 (sample point P
3 sample value) -1/4 (sample value of sample point P5)]-[-1/4 (sample value of sample point P2) +1/2 (sample value of sample point P3) -1/4 (sample value) P4 sample value)] ...
(1 ') The first term (in the first []) in this equation (1') represents the second-order derivative in the vertical direction of the image, and the second term (in the second []) is also horizontal. It represents the second derivative of the direction.

Therefore, if the luminance signal changes only in the vertical direction and the color does not change in both the horizontal and vertical directions on the sampled screen, the first term in the above equation is the sum of the contour component and the color signal component of the luminance signal. , The second term becomes a color signal component.

Therefore, according to the above equation (1 '), the color signal component included in the first term is canceled by the second term, and the contour component in the vertical direction with respect to the luminance signal can be extracted. If the luminance changes only in the horizontal direction and the color does not change in both the horizontal and vertical directions, the first term in equation (1 ') is the color signal component, and the second term is the contour component and color of the luminance signal. It is the sum of the signal components. Therefore, by changing the sign of the expression (1 '), the contour component in the horizontal direction with respect to the luminance signal can be extracted in the same manner as described above. Here, when the color signal spatially changes, the color signal component of the first term in the equation (1 ′) and the color signal component of the second term do not exactly match, and are therefore given by the equation (1 ′). Although some color signal component leakage occurs in the contour compensation signal, there is no practical problem.

Next, the control of the above-mentioned sign inversion will be described.

At time T, the output of the third delay circuit 21 is the sampling point P
This is a sampled value of 3 and this signal is doubled by the doubling circuit 29 and supplied to the second and third subtracting circuits 242 and 243. The output (P1 + P5) of the second addition circuit 222 is applied to the other input of the third subtraction circuit 243, and the output of the second addition circuit 222 is output from the output of the doubling circuit 29 (2 · P
The result obtained by subtracting 3) is the first absolute value circuit 251.
The absolute value is taken with. Therefore, the first absolute value circuit 251
Output signal T1 of T1 = | (sample value of sample point P1) -2 (sample value of sample point P3) + (sample value of sample point P5) | The output (P2 + P4) of the first addition circuit 221 is given to the other input of the second subtraction circuit 242,
The result obtained by subtracting the output (2 · P3) of the doubling circuit 29 from the output of the first adding circuit 221 takes the absolute value in the second absolute value circuit 252. Therefore, the output signal T2 of the second absolute value circuit 252 is T2 = | (sample value of sample point P2) -2 (sample value of sample point P3) + (sample value of sample point P4) |

The above-mentioned first absolute value circuit 251, second absolute value circuit 25
The respective output signals T1 and T2 of 2 are given to the comparison circuit 26, and their magnitudes are compared in the comparison circuit 26,
A control signal is sent to the sign inverting circuit 27 according to the comparison result. The sign inversion circuit 27 includes a first subtraction circuit 24.
When the output signals T1 and T2 are T1 ≧ T2, the comparison circuit 26 outputs T1 <T2 so that the output of the sign inversion circuit 27 becomes the output signal of the first subtraction circuit 241. In this case, the control signal is sent to the sign inverting circuit 27 so that the output of the sign inverting circuit 27 becomes a signal obtained by inverting the sign of the output signal of the first subtraction path 241. The output signal of the sign inversion circuit 27 is given to the multiplication circuit 28, and the output of the sign inversion circuit 27 is multiplied by N (N: real number) in the multiplication circuit 28. Although not shown, the multiplier N is externally controlled by, for example, a microcomputer.

According to this embodiment, since the contour signal is directly extracted from the composite video signal, the number of line delay circuits can be reduced and the cost can be reduced as compared with the conventional device. In addition, the detection of the image change direction is performed by comparing the magnitudes of the differential output values in the vertical and horizontal directions, that is, the difference between the sample values of the upper and lower sample points P1 and P5 close to the sample point P3, and the left and right sample points P2. , P4 as well as the difference between the signal changes including the sample value of the sample point P3 as well as the sample value of the sample point P3. Therefore, for example, on the upper sample point P1, the sample point P3, and the lower sample point P5. If there are thin lines,
The horizontal edge signal is reliably extracted without missing a steep horizontal signal change in which the left sampling point P2, the sampling point P3, and the right sampling point P4 are arranged, and the contour extraction is faithful. It is possible to obtain a good reproduced image.

Further, in the present embodiment, since the sample points whose color subcarrier phases are opposite to each other are located at the upper, lower, left and right positions with respect to the sample point of interest, the horizontal contour signal and the vertical contour signal can be simultaneously extracted.

〔The invention's effect〕

As described above, according to the present invention, the sample value of the sampling point and the samples of the upper and lower sampling points that are located above and below the sampling point and have the same color subcarrier phase and the opposite phase of the sampling point. The signal change amount in the vertical direction is detected from the value, and the sampled value of the sampled point and the left and right of the sampled point are in phase with each other in the color subcarrier phase and opposite to the sampled point.
Detect the amount of signal change in the horizontal direction from the sample value at the right sample point,
By comparing the magnitude of the signal change amount in the vertical direction and the signal change amount in the horizontal direction, one of the vertical and horizontal contour signals is extracted and output. Therefore, Y / C separation and contour extraction can be performed. In the circuit configuration that integrates both systems,
The line delay circuit required for the YC separation circuit and the line delay circuit required for the contour extraction filter can be shared, and the line delay circuit can be reduced.

Further, the detection of the image change direction is performed by comparing the magnitudes of the differential output values in the vertical and horizontal directions, that is, the difference between the sample values of the upper and lower sample points and the difference between the sample values of the left and right sample points that are close to the sample point. Not only that, but it is performed by comparing the magnitude of the signal change amount including the sample value of the sample point, so the situation of the sample point can be fed back and the contour extraction at the sample point can be performed more faithfully. effective.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of an NTSC adaptive contour extraction filter according to an embodiment of the present invention, and FIGS. 2 and 3 show a sampling signal sequence of an NTSC composite video signal sampled by sampling pulses. FIG. 4 is a schematic block diagram of a conventional horizontal contour extraction filter, FIG. 5 is a schematic block diagram of a conventional vertical contour extraction filter, and FIG. 6 is a schematic block diagram of a conventional contour extraction filter. In the figure, 1 is an input terminal, 2 is an A / D converter, 4 is an output terminal, 21, 22, 201 and 202 are delay circuits, 2
3, 29 are coefficient circuits, 26 is a comparison circuit, 27 is a sign inversion circuit, 28 is a multiplication circuit, 221 and 222 are addition circuits, 2
41 to 243 are subtraction circuits, 251 and 252 are absolute value circuits, and 11 is a contour extraction filter.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP 61-35085 (JP, A) JP 61-35086 (JP, A) JP 61-35087 (JP, A) JP 61- 189792 (JP, A) JP 62-7293 (JP, A) JP 59-40788 (JP, A) JP 60-241386 (JP, A)

Claims (1)

[Claims] 1. An NTSC composite video signal is input,
In an NTSC adaptive contour extraction filter for extracting the contour component of the composite video signal, A / D conversion means for sampling the composite video signal at a frequency synchronized with the horizontal scanning frequency and converting it into a digital signal, and digitized The sample value of the sample point of the composite video signal, and the sample values of the upper and lower sample points located above and below the sample point and having the same color subcarrier phase and the opposite phase of the sample point, Simultaneously extract the sample values of the left and right sample points, which are located on the left and right of the sample point on the same horizontal scanning line as the sample point and have the same color subcarrier phase as each other and the opposite phase of the sample point. And a first signal change amount detecting means for detecting a vertical signal change amount at the sample point by using the sample values of the sample point, the upper sample point, and the lower sample point. , Above A second signal change amount detecting means for detecting a signal change amount in the horizontal direction at the sample point by using sample values of the sample point, the left sample point, and the right sample point; An NTSC adaptive contour extraction filter, which is configured to extract and output one of a vertical direction and a horizontal direction contour signal by comparing the magnitude of a signal variation amount and a horizontal direction signal variation amount.