JPH0516708B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a video signal processing circuit, and in particular,
Video signal S/N using vertical correlation of images
The present invention relates to a signal processing circuit suitable for improving image sharpness.

[Background of the invention]

There is a so-called comb filter that utilizes the line correlation of an image signal as a means for improving the S/N of a video signal and improving the sharpness of an image. Figure 1 shows a comb filter that adds the signals before and after a 1-line delay line that delays the signal by 1 horizontal scanning time and outputs the result. Signals with line correlation are multiplied by 2, and random noise is multiplied by √2. This is an S/N improvement circuit that improves the S/N ratio by a factor of √2 (3 dB). However, this circuit naturally has the disadvantage that the vertical contours of the image become blurred because the signals of the previous line are added. Next, Figure 2 shows
A comb filter that adds an appropriate amount of the difference between the signals before and after a 1-line delay line to the original signal, and improves image sharpness by emphasizing signals that have no line correlation (signals in areas with contours in the vertical direction of the image). improve. However, this circuit has the drawback of emphasizing random sounds in a flat image portion with no outline in the vertical direction, thereby worsening the S/N ratio.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned drawbacks of the prior art and to provide an S/N improvement circuit with less blurring of image contours and an image sharpness improvement circuit with less deterioration of S/N.

[Summary of the invention]

The first main point of the present invention to achieve the above object is to
The output signal of a contour signal extraction type comb filter that extracts the upper and lower contours of an image by subtracting the signals before and after the line delay line, and the output signal of the comb filter are transferred to a limiter circuit or a base clip that prevents small amplitude signals from passing through. By adding or subtracting an appropriate amount of the signal obtained through the circuit to the original signal, it acts as an S/N improvement circuit for the signal of the image part where there is no or weak upper and lower contours of the image, and improves the signal of the image part with strong contours. The purpose is to act on the signal as an image sharpness improvement circuit.

The second point is to make the limiter level or clip level of the limiter circuit or base clip circuit variable, and to adjust this in conjunction with AGC (automatic gain control) in a system in which the circuit is incorporated. Either the noise level is automatically controlled so that it is large when the gain is high and it is small when the gain is low, or it is controlled by the output signal of a circuit that detects the amount of noise in the video signal to be processed.

The third point is that the mixing ratio at which the output signal of the comb filter and the signal obtained through the limiter circuit or the base clip circuit are mixed into the original signal is controlled in conjunction with the AGC or in the signal to be processed. It is controlled by the output signal of a circuit that detects the amount of noise.

[Embodiments of the invention]

Examples of the present invention will be described below. FIG. 3 shows a first embodiment of the present invention, in which 1 is an input terminal for a video signal to be processed, 2 is an output terminal, 4 is a contour signal extraction type comb filter for extracting the upper and lower contours of an image, and 5 is a comb filter for extracting upper and lower contours of an image.
1 is a limiter circuit or a base clip circuit for blocking the passage of small amplitude signals, 6 and 7 are coefficient circuits that determine the mixing ratio of signals, and 8 is a mixing circuit. First, the operation will be described in the case where the concrete circuit of the contour signal extraction type comb filter is the circuit shown in FIG. 4, and 5 is a limiter circuit having the characteristics shown in FIG. The limiter level of the limiter circuit is set to be approximately equal to or slightly larger than the noise level in the signal input from the contour signal extraction type comb filter 4 to the limiter circuit. The input signal is displayed as S, the signal after one line delay as Sd, and the output signal as So. The input signal of the limiter circuit 5 is expressed as (S-Sd), and in the part where there is no vertical contour in the image, (S
−Sd) does not reach the limiter level and passes through the limiter circuit as it is, so the output signal So
is expressed as:

So={1+(α+β)}S−(α+β)Sd (1) Here, if we set the condition that the coefficients of S and Sd are equal, i.e., α+β=−1/2 (2), this is one line delay. The signals before and after the line are added in equal amounts, and the circuit functions as an S/N improving circuit equivalent to the circuit shown in FIG.

Next, in areas where the image has vertical contours,
Since the amplitude of the limiter circuit input signal represented by (S-Sd) is large enough to exceed the limiter level, passage through the limiter circuit is blocked, and the output signal So is represented by the following equation.

So=S+α(S-Sd) (3) Since the signal represented by (S-Sd) represents the contour of the screen in the vertical direction, by setting α to a positive value, the output signal So becomes , contour signals are added to produce an edge enhancement signal, and this circuit functions as a sharpness improvement circuit. Specific numerical examples of α and β are α=1/2 and β=−1. As a result, an image signal can be obtained in which the S/N ratio is improved in areas where the image does not have upper and lower contours, and the sharpness is improved in areas where there are contours. As yet another example of a combination of α and β, there is a method of setting them to values within the range defined by equation (4) -1/2<α+β<0 (4). In this case, the S/N improvement effect of a portion without an outline is lower than that when the condition of equation (2) is set, but the deterioration of minute changes (details) within the flat portion of the image is reduced accordingly.

Next, the operation will be described in the case where the circuit block indicated by 5 in FIG. 3 is a base clip circuit having the characteristics shown in FIG. It is assumed that the clip level of the base clip circuit is set approximately equal to or slightly higher than the noise level in the signal input to the base clip circuit. In areas where there is no vertical contour in the image, the input signal (S-Sd) to the base clip circuit does not reach the clip level and is blocked from passing through the base clip circuit.
The output signal So at this time is expressed by the following equation.

So=(1+α)S−αSd (5) Here, if we set the condition that the coefficients of S and Sd are equal, that is, α=-1/2 (6), this means that the signals before and after the 1-line delay line By adding equal amounts, the circuit shown in FIG. 1 acts as an equivalent S/N improving circuit.

Next, in areas where the image has vertical contours,
Since the amplitude of the base clip circuit input signal represented by (S-Sd) exceeds the clip level and passes through the base clip circuit, the output signal So is represented by the following equation.

So=S+(α+β)(S-Sd) (7) The signal represented by (S-Sd) is a signal representing the contour of the screen in the vertical direction, so it satisfies equation (8). (α+β)> 0 (8) When α and β are determined, the output signal So becomes an enhanced signal by adding the contour signal, and this circuit acts as a sharpness improvement circuit. Specific numerical examples of α and β are α=−1/2 and β=1.

This combination performs a circuit operation completely equivalent to the circuit operation obtained by the combination of α=1/2 and β=-1 when using the limiter circuit described above. This is natural since limiter circuits and base clip circuits have a complementary relationship in which the characteristics obtained by subtracting the output signal of one circuit from the input signal are equivalent to the characteristics of the other circuit. It is about. Therefore, as in the above description, an S/N improvement effect can be obtained in a portion without an image contour, and a sharpness improvement effect can be obtained in a portion with an image contour.

Next, a second embodiment of the present invention will be described using FIG. 7. In this embodiment, the image contour signal extraction comb filter 4 consists of two line delay lines. The input signal is applied to terminal 1, but in order to better understand the correspondence with the first embodiment, terminal 1' will be regarded as a virtual input terminal and the signal at that point will be referred to as S. do. And terminal 1 has terminal 1'
It is assumed that a signal that is one line time ahead of the signal is generated, and this is expressed as Sa. Therefore,
The output signal of the contour signal extraction type comb filter 4 in this embodiment is expressed as S-1/2 (Sa+Sd).

This is a contour signal that occurs oddly symmetrically at the contour in the vertical direction. In the image portion where the contour signal does not reach the limiter level of the limiter circuit 5, the output signal So is expressed by the following equation.

So={1+(α+β)}S-1/2(α+β)(Sa+
Sd) (9) Here, if we set the condition that the coefficient ratio of S and (Sa+Sd) is 2:1, that is, (α+β)=-1/2 (10), So becomes So=1/2S+1/4( Sa + Sd) (9)′ (9)′ Since it is a signal that is a mixture of a signal that is one line ahead of time and a signal that is delayed, the random noise is different from the original signal. reduced to

Next, in areas where the image has vertical contours,
Since the amplitude of the contour signal exceeds the limiter level of the limiter circuit and is prevented from passing through the limiter circuit, the output signal
So is expressed by the following formula.

So=S+α{S-1/2(Sa+Sd)} (11) Since the signal represented by {S-1/2(Sa+Sd)} is a contour signal, by setting α to a positive value, the contour This results in a video signal with enhanced emphasis. Comparing this example with the first example, the S/N
The improvement effect is large and the contour is emphasized with odd symmetry (pre-shoot and overshoot). Note that in this embodiment, as in the case of the first embodiment, it is possible to replace the circuit block 5 with a base clip circuit and to change equation (10) to equation (4). .

Next, a third embodiment of the present invention will be described using FIG. 8. The image contour extraction type comb filter 4 in this embodiment is a feedback type comb filter that feeds back a signal of the difference between the signals before and after a one-line delay line to the input at a feedback rate K. The transfer function Gk of this feedback comb filter is expressed by the following equation.

Gk=-1/1+k×1-ε ^-j 〓 ^t /1-k/1+ ^kε-
j 〓 ^t (12) Here, T is the delay time of one line delay line. The frequency function of the absolute value of Gk |Gk| is illustrated in FIG. 9A, taking the case of K=1 as an example. In the figure, f _H is the horizontal scanning repetition frequency and is equal to the reciprocal of the delay time T of one line delay line. Gk has a zero response at frequencies that are integral multiples of _fH . This means that the video signal passed through this comb filter contains only the contour signal components in the vertical direction of the image. Therefore, the circuit 4 within the dotted line frame in FIG. 8 functions as an image contour extraction type comb filter. In areas where there is no vertical contour in the image, the amplitude of the output signal of the image contour extraction type comb filter is small and the limiter circuit 5
Because it passes through this without reaching the limiter level of
Transfer function Gp from terminal 1 to terminal 2 at this time
is expressed by the following formula.

Gp=1/1+k・1+{k-((α+β)}-{k-
(α+β)}ε ^-j 〓 ^t /1-k/1+kε ^-j 〓 ^t (13) Here, if (α+β)=K (14) is set, equation (13) becomes Gp=1/1+k・1/1 −k/1+kε ^−j 〓 ^t (13)′. Also, if (α+β)=K+1/2 (15) is set, Gp=1/1+k・1/2・1+ε ^-j 〓 ^t /1−k/1
+kε ^-j 〓 ^t (13)″.The frequency function of the absolute value of Gp |Gp| is K=
1 is shown in FIG. 9B as an example. It has a peak at a frequency that is an integer multiple of _fH , and reaches a minimum at a frequency that is shifted by 1/2 _H from the integer multiple frequency. Since the frequency spectrum of the video signal in the part of the image that has no outline in the vertical direction concentrates around an integer multiple of _fH , there is little image deterioration due to passing it through this circuit, and noise in the frequency domain shifted by 1/ _2fH is suppressed. Only S/N is suppressed.
is improved. Note that the values of K and (α+β) are not limited to the conditions of equations (14) and (15),
It can take any value in its vicinity.

Next, in areas where the image has vertical contours,
Since the output signal of the image contour extraction comb filter has a large amplitude and is prevented from passing through the limiter circuit, Gp at this time is expressed by the following equation.

Gp=1/1+k・(1+k−α)・1−k−α/1+
k-αε ^-j 〓 ^t /1-k/1+kε ^-j 〓 ^t (16) Here, in the setting of α<0, k-α/1+k-α>k/1+k (17) holds, and in this case As shown in FIG. 9C, Gp exhibits a characteristic in which the frequency region shifted by 1/2f _H is peaked, and this circuit functions as an image contour emphasizing circuit.

As described above, in this embodiment, a circuit is obtained which functions as an S/N improvement circuit in areas where there are no upper and lower contours in the image, and as a sharpness improvement circuit in areas with contours. The feature of this embodiment is that there is a degree of freedom in setting the maximum value of the S/N improvement effect by setting the feedback rate K, and a greater effect than that of the previous embodiment can be obtained.

In this embodiment, as in the case of the first embodiment, the circuit block 5 can be constructed by using a base clip circuit instead of a limiter circuit, but detailed explanation will be omitted.

Next, a fourth embodiment of the present invention will be described using FIG. 10. FIG. 10 is a circuit diagram of an image contour extraction type comb filter used in this embodiment.
The difference from that used in the third embodiment (FIG. 8) is the injection point of the feedback signal. The transfer function from terminals 41 to 42 of this circuit is expressed by the following equation.

Gk=1−ε ^-j 〓 ^t /1−k'ε ^-j 〓 ^t (18) Comparing equation (18) with equation (12) of the transfer function of the comb filter in the third embodiment, we get , K'=k/1+k (19) When setting K'=k/1+k (19), a transfer function equivalent to (1+K) times (=k/k' times) the transfer function of the comb filter of the third embodiment can be obtained. I understand. The frequency function of |Gk| is shown in FIG. 11, taking the case of K'=1/2 as an example. In this way, the circuit shown in FIG. 10 provides a contour signal K/K' times that of the third embodiment. In this embodiment using this, by setting α and β in the third embodiment to α and β multiplied by K′/K, the operation and effects completely equivalent to those in the third embodiment can be obtained. It will be done.

Next, a fifth embodiment of the present invention will be described using FIG. 12. In this embodiment, 9 is a variable gain circuit, 10 is a control circuit for controlling the gain of the variable gain circuit, and the others are the same as in the first embodiment, but 5 is a variable gain circuit with a variable limiter level or clip level. It is a limiter circuit or a base clip circuit, and the coefficients 6 and 7 of α and β are variable under the control of the control circuit 10. The variable gain circuit 9 is usually an AGC (Automatic Gain
When the input signal is insufficient in magnitude, it increases the gain and maintains the output signal level at an appropriate level. However, when the input signal decreases, the S/N usually deteriorates, and in such a case, the S/N according to the present invention
It is effective to strengthen the improving effect and weaken the sharpness improving effect. In this embodiment, first, the limiter level (or clip level) of the limiter circuit (or base clip circuit) is increased in conjunction with the increase in the gain of the variable gain circuit 9. Since it acts as an S/N improvement circuit in the portion where the contour signal does not reach the limiter level (or clip level), increasing the limiter level widens the range in which it acts as an S/N improvement circuit, increasing the S/N improvement effect. Second, the coefficients of α and β are controlled in a direction that increases the S/N improvement and weakens the sharpness improvement in conjunction with the gain increase of the variable gain circuit. In the circuit format of the first embodiment using a limiter circuit, |α| is decreased and β is controlled to approach −1. According to the fifth implementation, the degree of S/N improvement and sharpness improvement is optimally controlled depending on when the S/N of the input signal is good and bad. Mutually contradictory side effects can be suppressed to a minimum, and good images without deterioration in image quality can be obtained.

Next, a sixth embodiment of the present invention will be described using FIG. 13. In this embodiment, 10' is the limiter level (or clip level) of the limiter circuit (or base clip circuit), and the coefficient α of the coefficient circuit.
A control circuit 11 controls β, and a noise amount detection circuit 11 detects the amount of noise in the input signal. The rest is the same as the first embodiment. The control circuit 10' detects the amount of noise in the input signal and adjusts the limiter level (clip level) and α, according to the detected signal.
Control β. When the amount of noise is large, the limiter level and α and β are controlled in the same way as when the gain of the variable gain circuit in the fifth embodiment is high, and the S/N and sharpness are improved by an appropriate amount commensurate with the amount of noise. Close. An example of the noise detection circuit 11 extracts the horizontal blanking or vertical blanking portion of the input video signal through a gate circuit, amplifies the noise included therein, and then detects the noise to reduce the amount of noise. This is a method to obtain the appropriate DC voltage. Another example is when the present invention is implemented in a television camera, extracting a no-signal part of a signal waveform called optical black, which is usually used as a reference for clamping, and detecting the amount of noise superimposed thereon. There is a way. According to this embodiment, a system in which the present invention is implemented is provided with a system similar to that of the fifth embodiment.
Even when not equipped with an AGC circuit, S/
The effects of N improvement and sharpness improvement can be automatically set to optimal amounts. Therefore, it is effective even for signals such as TV broadcast reception signals and VTR playback signals, where the amount of noise depends largely on the signal generator.

〔Effect of the invention〕

As explained above, according to the present invention, in a portion without an image contour, it acts as an S/N improvement circuit,
By acting as a sharpness improvement circuit that does not stop the S/N improvement operation in a portion where there is an image contour, S/N improvement with less blurring of the contour and sharpness improvement with less S/N deterioration can be achieved. In addition, the transition point and amount of improvement of the above two effects are determined by the AGC of the system.
Since the control is performed in conjunction with the amount of noise in the input signal or in response to the amount of noise in the input signal, the above effect is greatly enhanced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a conventional S/N improvement circuit, FIG. 2 is a block diagram showing an example of a conventional sharpness improvement circuit, and FIG. 3 is a block diagram showing an example of a conventional sharpness improvement circuit. FIG. 4 is a block diagram showing an example of the contour extraction type comb filter used in this embodiment, and FIGS. 5 and 6 are characteristic diagrams showing examples of the characteristics of the limiter circuit and the base clip circuit, respectively. , 7th
The figure is a block diagram showing a second embodiment of the present invention.
FIG. 8 is a block diagram showing a third embodiment of the present invention, FIG. 9 is a characteristic diagram for explaining the operation of the third embodiment, and FIG. 10 is a fourth embodiment of the present invention. FIG. 11 is a characteristic diagram for explaining the operation of the fourth embodiment, and FIG. 12 is a characteristic diagram for explaining the operation of the fourth embodiment.
FIG. 13 is a block diagram showing a sixth embodiment of the present invention. 1... Input terminal, 2... Output terminal, 3... 1-line delay line, 4... Image contour extraction type comb filter, 5... Limiter circuit or base clip circuit,
6, 7... Coefficient circuit, 8... Mixing circuit, 9... Variable gain circuit, 10, 10'... Control circuit, 11...
...Noise amount detection circuit.

Claims (1)

[Claims] 1. A contour extraction type comb filter which uses a video signal as an input signal and extracts the upper and lower contours of an image by subtracting signals before and after a line delay line; and an output signal of the contour extraction type comb filter. A limiter circuit with a variable limiter level or a base clip circuit with a variable clip level as an input signal, and both the output signal of the contour extraction comb filter and the output signal of the limiter circuit or base clip circuit as the video signal. A video signal processing circuit comprising: a mixing circuit that performs addition or subtraction mixing in an appropriate amount; 2. A patent claim characterized in that the limiter level or clip level of the limiter circuit with a variable limiter level or the base clip circuit with a variable clip level is controlled in conjunction with an AGC circuit of a system including the video signal processing circuit. The video signal processing circuit according to item 1. 3. The limiter circuit with a variable limiter level or the base clip circuit with a variable clip level is such that the limiter level or clip level corresponds to at least one waveform portion of horizontal blanking, vertical blanking, or optical black in the video signal to be processed. 2. The video signal processing circuit according to claim 1, wherein the video signal processing circuit is controlled by a control signal obtained by detecting superimposed noise. 4. The mixing ratio of the mixing circuit is controlled by a control signal obtained by detecting noise superimposed on at least one waveform portion of horizontal blanking, vertical blanking, or optical black in the video signal to be processed. The video signal processing circuit according to any one of claims 1 to 3, characterized in that: 5 A system including the above video signal processing circuit.
The video signal processing circuit according to any one of claims 1 to 3, characterized in that it is controlled in conjunction with an AGC circuit.