JPH10150582A - Contour correction circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the contour correction circuit, in which no overshoot/ preshoot exists in the signal and no signal amplitude is deteriorated, even when the frequency of the signal is high. SOLUTION: An input video signal (a) is given to a 2nd order differentiation circuit 11, in which a 2nd order differentiation signal (b) is generated and the signal (b) is delayed by times t1, t2 to obtain 2nd order differentiation signals c, d. The 2nd order differentiation signals b-d are given to a nonlinear filter 14, a correction signal (e) is outputted from the output of the filter 14, the signals a, e are added to provide an output of an output image signal (f) from which preshoot/overshoot is eliminated. Thus, the nonlinear filter 14 is acted on only the 2nd order derivative signal (b) with respect to the input signal (a) to obtain the correction signal (e) and the correction signal (e) is added to the input signal (a), then the output signal (f) without overshoot/preshoot is obtained. Also since the input signal (a) is not given to the nonlinear filter 14, a good response is obtained even when the circuit receives a signal with a high frequency.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a contour correction circuit having little overshoot / preshoot in an image signal and good response to a high-frequency signal.

[0002]

2. Description of the Related Art Conventionally, various circuit systems have been studied and implemented for the purpose of improving image quality. A general second-order differential type contour correction circuit is shown in FIG. 9 and will be described with reference to FIG. 10 showing signal waveforms of main parts of this figure.

FIG. 9 shows an example in which the input image signal a has a rising waveform as shown in FIG. The input signal a is secondarily differentiated by a second differentiating circuit 91 to obtain a signal b. The signal b is a second-order differentiated high-frequency component, which is subjected to noise processing by a coring circuit 92 to obtain a correction signal.

The output signal d is a signal having overshoot and preshoot as shown in FIG. Increasing the level of the correction signal applied to the signal c increases the correction effect, but also increases the levels of overshoot and preshoot, so that the level of the outline of the image rises, giving rise to a ghost-like feeling, and causing image quality deterioration. .

[0005] The overshoot / preshoot problem is as follows.
It can be removed by using a non-linear circuit.
For example, there is Japanese Patent Publication No. 1-4548. FIG. 11 is a circuit block diagram, and FIG. 12 is a waveform diagram.

The input image signal a has a rising step waveform as shown in FIG. The input signal a is delayed by the delay circuit 111 for a time t1 to obtain a signal b. Similarly, the input signal a is converted to the signal c by the delay circuit 112 for the time t2.
Get. From the delay signal b, the second
The second derivative signal is obtained, and the minute level signal is removed by the coring circuit 114 to obtain a signal d from which noise has been removed. The signal d and the signal b are added by the adder 115 to obtain a signal e. This circuit is similar in principle to the contour correction circuit using the second derivative shown in FIG.

On the other hand, the maximum value f is selected from the input signal a and the signal c by using the maximum value circuit 116 and the minimum value circuit 117 is selected.
Is used to select the minimum value g. The signals e, f, and g are selected by the median selection filter 118 to select the data at the middle level among the three data, and the selected data is set as the output h.

The signal waveforms indicated by e, f, and g in FIG. 12 are signals e, f, and g. The output signal h of the median selection filter 118 selects the median value. The signal e is the same as the output signal of FIG.
There is preshoot. The output signal h is the output of the median selection filter 118 from which these shoots have been removed. For this reason, although the method uses the high-frequency second-order differential signal, the image quality is less deteriorated because overshoot and preshoot can be removed.

However, there is a problem when the overshoot / preshoot is removed by using this circuit. Since this circuit performs nonlinear processing, the output response differs depending on the waveform of the input signal. In particular, this problem will be described by taking, as an example, a case where a high frequency signal is input.

That is, although waveform diagrams in the middle are omitted, FIG. 13A shows signals e, f, and g input to the median selection filter 118. The signal e has a large amplitude for high-frequency emphasis in contour correction. A high signal means that the delay time of the delay times t1 and t2 approaches the signal period.

Accordingly, the signals e and e shown in FIG.
As a result of selecting the median values of f and g, the peak value of the sine wave decreases, and the total signal amplitude decreases. This is because nonlinear processing is performed on the corrected signal.

[0012]

The above-described conventional contour correction circuit capable of removing the overshoot and preshoot has a problem that the signal amplitude is deteriorated for a high-frequency signal.

The present invention provides a contour correction circuit that has no overshoot / preshoot and does not degrade the signal amplitude even for a high-frequency signal.

[0014]

In order to solve the above-mentioned problems, in the contour correction circuit of the present invention, an input video signal is input to a second-order differentiation circuit to generate a first second-order differentiation signal,
The first and second secondary differential signals delayed by t1 and t2 are used as second and third secondary differential signals, and the first to third secondary differential signals are input to a nonlinear filter. By outputting a correction signal and adding the correction signal to the input video signal, it is possible to output an output image signal from which preshoots and overshoots have been removed.

By the above means, a correction signal is obtained by applying a non-linear filter only to the second derivative signal of the input signal. By adding this correction signal to the input signal, an output signal having no overshoot / preshoot can be obtained. In addition, since a non-linear filter is not applied to the input signal, the response is good even when a high frequency signal is input.

[0016]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram for explaining an embodiment of the present invention, which will be described together with FIG. 2 showing signal waveforms of main parts in FIG.

In FIG. 1, the input video signal is
The case where the step signal is as shown in FIG. The input signal “a” obtains a differential signal “b” through the second differentiating circuit 11. The signal c and the signal d are obtained from the signal b via the delay circuit 12 for delaying the time t1 and the delay circuit 13 for delaying the time t2. The delay time is t
It is general to set 1 = t2. These signals b ~
d is combined with the positive and negative processing circuits 141 and 142.
The nonlinear processing is performed by the nonlinear filter 14 configured by the selection circuit 143.

On the other hand, the primary differential signal g is obtained from the input signal a by the primary differential circuit 15, and the sign of the signal is obtained as the signal h via the sign extracting circuit 16. In FIG. 2, the signal h indicates positive. In this case, zero is determined to be positive, but there is no problem on the function even if it is determined to be negative. This is because the first and second derivative signals of the flat portion are zero. The obtained correction signal e is input to one input of an adder 18 as a signal f from which the minute amplitude has been removed by the coring circuit 17. A signal f and an input signal a
And a signal j obtained through a time adjusting circuit 19 for adjusting the time. From the output of the adder 18, an output signal i of a video whose contour has been corrected is obtained.

The waveform diagram of FIG. 3 shows the signal processing on the positive polarity side of the nonlinear filter 14. By taking the minimum value of the signals obtained by clipping the negative sides of the c and d signals, the correction signal e1 in the left half of the transient portion of the rising waveform is obtained.
Get. Similarly, by clipping the positive sides of the b and c signals in the negative polarity processing and taking the maximum value, a correction signal e2 in the right half of the transient portion of the rising waveform is obtained. The two signals e1 and e2 are combined and selected to obtain the correction signal e of FIG. 2, and the correction signal e is input to the coring circuit 17, where the signal f from which the minute amplitude has been removed can be obtained.

In this embodiment, the correction signal e is applied to only the second derivative signal of the input signal a by applying the nonlinear filter 14.
Get. The adder 18 converts the correction signal e into the input signal a
, An output having no overshoot or preshoot can be obtained. Further, since the non-linear filter 14 is not applied to the input signal a, the response is good even for a high frequency signal.

Next, a more specific configuration of the nonlinear filter 14 will be described with reference to the block diagram of FIG. The signal b of the secondary differentiating circuit 11 is input to the negative clip circuit 1411 and the positive clip 1418, respectively. Delay circuit 1
The signal c of 2 is input to the negative clip circuits 1412 and 1416 and the positive clip circuits 1413 and 1417, respectively. The signal d of the delay circuit 13 is
4 and the negative side clipping circuit 1415, respectively.

The minimum value of the output signals of the negative clip circuits 1411 and 1412 is extracted by the minimum value circuit 1421 and input to one of the adders 1431. Positive clip circuit 14
The maximum value of the output signals of the
At 22, it is taken out and inputted to the other end of the adder 1431. The minimum value of the output signals of the negative clip circuits 1415 and 1416 is extracted by the minimum value circuit 1423 and input to one of the adders 1432. Positive clip circuit 1417, 1418
The maximum value of the output signal is taken by the maximum value circuit 1424 and input to the other end of the adder 1432.

The outputs of the adders 1431 and 1432 are input to a selection circuit 1433, respectively. The output of the selection circuit 1433 is selected based on the signal h obtained by extracting the sign of the primary differential signal g of the primary differentiating circuit 15 using the code extracting circuit 16.

Here, following the description of FIG. 3, the processing will be described on the positive side of FIG. That is, the signals b and d in FIG. 2 are input to the negative side clipping circuits 1415 and 1416, and when the negative side is clipped, the signal becomes c + d in FIG. When the signal passes through the minimum value circuit 1423, the signal of e1 in FIG. 3 can be obtained. Also, the signals b and c in FIG. 2 are input to the positive side clipping circuits 1417 and 1418, and when the positive side is clipped, the signal becomes b + c in FIG. The signal is sent to the maximum value circuit 14.
When the signal passes through 24, the signal of e2 in FIG. 3 can be obtained.

The signals e1 and e2 are synthesized by an adder 1432, and the output signal h of the code extraction circuit 16 selects the selection circuit 14
If the positive side is selected by 33, the signal e in FIG.

Here, the processing on the positive side has been described, but the same applies to the signal processing on the negative side, and a detailed description thereof will be omitted. The nonlinear filter 14 clips the positive side and the negative side of the signals b, c, and d, and combines and selects the maximum value and the minimum value from each clipped signal.

Another specific configuration example of the nonlinear filter 14 will be described with reference to FIG. This specific example is obtained by removing the circuit that can be shared from the circuit of FIG. That is, a selection circuit 52 is provided before the negative clipping circuit 51 that clips the negative sides of the signals b and d, and a selection circuit 54 is provided before the positive clipping circuit 53 that clips the positive sides of the signals b and d. The selection circuits 52 and 54 are selected based on the sign signal h of the first derivative.

As described above, the selection circuits 52 and 54 are simply arranged at the preceding stage of the negative and positive clip circuits 51 and 53, and
This is configured to be selected by the sign signal h of the next differentiation. In this case, the number of clip circuits, maximum value circuits, and minimum value circuits becomes half each while having the same functions as in FIG. 4, and a simple circuit can be realized.

FIG. 6 is a block diagram for explaining still another specific configuration of the nonlinear filter 14. As shown in FIG. In this specific example, the maximum value / minimum value circuits 61 and 62 are arranged in front of the positive and negative clip circuits 63 and 64, and further, a selection circuit is provided in front of the maximum value / minimum value circuits 61 and 62. 65 and 66 are arranged.

The non-linear filter of this specific example has the same function as that of FIG. 4 but can further eliminate one clip circuit on the negative side and one clip circuit on the positive side as compared with FIG.

FIG. 7 is a block diagram for explaining another embodiment of the present invention. The same functions as those in FIG. 1 are denoted by the same reference numerals, and different portions will be mainly described here.

That is, in this embodiment, a signal obtained from the absolute value circuit 71 and the comparator 72 from the output signal g of the primary differentiating circuit 15 detects whether or not the image is a flat portion. The flat portion has neither the first derivative signal nor the second derivative signal. However, since the noise component is present, if the noise component is added to the main signal, the S / N ratio of the output signal becomes worse. When the flat portion of the image is detected by the absolute value circuit 71 and the comparator 72, the level attenuating circuit 73 reduces the level of the output signal f of the coring circuit 17, which is the correction signal, or sets it to zero, thereby reducing the deterioration of the SN ratio. Can be prevented.

The position of the level attenuating circuit 73 may be anywhere on the circuit diagram as long as it functions the same. For example, it may be at the input stage of the nonlinear filter 14,
It may be provided at the output stage of the secondary differentiating circuit 11.

According to the present invention, as shown in FIG. 8, even if the signal j obtained by passing the input signal a through the time adjusting circuit 19 is a high frequency sine wave as shown in FIG.
Since the high-frequency sine wave component is stored in the signal i, which is the output of 8, the response in the high frequency range is good and the deterioration of the signal amplitude can be prevented.

[0035]

As described above, according to the contour correction circuit of the present invention, the contour correction without overshoot and preshoot can be performed, the correction level can be increased, and a sharper and higher quality image can be obtained. Can be Further, since the main signal does not pass through the non-linear filter, even if a high frequency signal is input, the main signal is stored as it is, so that the signal amplitude does not deteriorate.

[Brief description of the drawings]

FIG. 1 is a block diagram for explaining an embodiment of the present invention;

FIG. 2 is a signal waveform diagram for explaining the operation of FIG.

FIG. 3 is a signal waveform diagram for explaining the operation of FIG. 1;

FIG. 4 is a block diagram for explaining a specific example of the nonlinear filter of FIG. 1;

FIG. 5 is a block diagram for explaining another specific example of the nonlinear filter of FIG. 1;

FIG. 6 is a block diagram for explaining another specific example of the nonlinear filter of FIG. 1;

FIG. 7 is a block diagram for explaining another embodiment of the present invention.

FIG. 8 is a signal waveform diagram for explaining the effect of the present invention.

FIG. 9 is a block diagram for explaining a conventional contour correction circuit.

FIG. 10 is a signal waveform diagram for explaining the operation of FIG. 9;

FIG. 11 is a block diagram for explaining another conventional contour correction circuit.

FIG. 12 is a signal waveform diagram for explaining the operation of FIG. 10;

FIG. 13 is a signal waveform diagram for explaining a conventional problem.

[Explanation of symbols]

11: second-order differentiator circuit, 12, 13: delay circuit, 14: nonlinear filter, 15: first-order differentiator circuit, 16: sign extraction circuit, 17: coring circuit, 18: adder, 19: time adjusting circuit.

Claims (5)

[Claims] 1. An input video signal is input to a secondary differentiating circuit to generate a first secondary differential signal, and the first secondary differential signal delayed by t1 and t2 is converted to second and third signals. By inputting the first to third secondary differential signals to a non-linear filter, outputting a correction signal from its output, and adding the correction signal to the input video signal, A contour correction circuit capable of outputting an output image signal from which preshoots and overshoots have been removed. 2. The method according to claim 1, wherein the nonlinear filter includes the first to third filters.
2. The contour correction circuit according to claim 1, wherein the processing is changed depending on the positive or negative polarity of the second derivative signal of the first and second signals. 3. The non-linear filter according to claim 1, wherein a signal subjected to signal processing on the positive polarity side and a signal on the negative polarity side is selectively output based on the polarity of a primary differential signal of the input video signal. The contour correction circuit as described. 4. The non-linear filter uses a signal obtained by clipping the negative side of the first to third differential signals, and selectively outputs the minimum value of at least two of the first to third differential signals. Positive signal processing, and negative signal processing for selecting and outputting the maximum value of at least two of the first to third differential signals using a signal obtained by clipping the positive side of the first to third differential signals. And a contour correction circuit comprising: 5. The method according to claim 1, wherein the correction signal added to the input video signal is reduced in amplitude by using amplitude information of a primary differential signal of the input video signal, or the correction signal is not output. The contour correction circuit according to claim 1, wherein