JPH0898058A - Contour emphasis device - Google Patents

Abstract

PURPOSE: To attain accurate contour emphasis objectively by providing a noise measurement circuit and a contour adjustment circuit adjusting a degree of contour emphasis in a signal corresponding to a composite video signal based on an output of the noise measurement circuit. CONSTITUTION: A luminance signal is inputted to an input 1 added to a line La and the waveform of the signal is corrected by sharpening a waveform slope part or the like. The signal at the input 1 is given to an input terminal of an adder 11 and an HPF amplifier (contour adjustment circuit) 12 comprising a transversal filter, in which horizontal and vertical two-dimension harmonics. A composite video signal being a noise detection object signal fed to a line Lc in a noise measurement circuit 14 is fed to a subtractor and a delay circuit and a difference signal between the input signal from the line Lc and a signal of one preceding field is formed. The signal is fed to the circuit 12 via a line Ld from the circuit 14 and the degree of contour emphasis in the horizontal direction is controlled to obtain optimum image quality in response to the noise amt. included in the signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a television (TV).
The present invention relates to a high-performance image quality improvement and contour enhancement device used in various video equipment such as a receiver and a video tape recorder (VTR).

[0002]

2. Description of the Related Art A conventional contour emphasizing apparatus will be sequentially described below with reference to FIG. Here, the target signal for contour enhancement is, for example, a luminance signal of the NTSC system or the like. In FIG. 6, block 61 is an adder (ADD), 62 is a high-pass filter (HPF), and 63 is an amplifier (AMP).
Is. Similarly, the line La is a signal input line, the line Lb is a signal output line, and the line Ld is an amplifier 63.
This is an input line for adjusting the amplification factor of.

From the line La, a luminance signal (input 1) having a spectral component in the range up to the frequency μm (about 4 MHz) is input as shown in FIG. 9 (a). Line La
The input luminance signal from is applied to one input terminal of the adder 61, and is also applied to the high-pass filter 62 at the same time. This high-pass filter 62 has a horizontal high-frequency signal component (see FIG. 9C).
Primary contour enhancement), or horizontal and vertical signal components (see FIG. 9).
(C) and FIG. 10 (e), this is a filter for extracting two-dimensional contour enhancement, and forms a so-called second-order differential waveform for contour enhancement.

The next block 63 is an amplifier, which serves to amplify the high frequency signal component supplied from the preceding stage.
The high-frequency signal component formed in this way becomes a signal for contour enhancement. From the input line Ld, this amplifier 6
A control signal for adjusting and controlling the amplification factor of No. 3 is added, and it is set so that a proper edge enhancement effect can be obtained. Since the sensitivity to contour enhancement differs among individuals, it is usually performed by manual control.

The output of the amplifier 63 is the adder 6 described above.
1 is added to the other input terminal and is added to the input signal (input 1). The line Lb is a signal output line from which a contour-emphasized luminance signal is output. As a result, when the input signal has a sloped waveform portion such as g2 shown in FIG. 10 (g), it is converted into a waveform having a steep edge like g1 in FIG.

The function of the conventional contour emphasizing device as shown in FIG. 6 is to make the corrugated slope steep in this way. Although the external control is possible in the conventional example, it is the same as it is if it is not adjusted by a person, so even if it is optimal in the receiving state of a certain channel (CH), another CH, especially a channel to which a noisy signal is input, Since the contour enhancement is originally enhancement of the high-frequency signal component, the high-frequency component of noise is also emphasized, and there is a defect that an image on which rough noise is superimposed is seen.

At this time, if the person watching the image is concerned about noise, it is necessary to make adjustments for optimization each time. Since ordinary people who usually watch TV do not make such adjustments on a small scale, there is a drawback in that they end up looking at the image in a deteriorated state. In general feeling, it is necessary to increase the degree of contour enhancement when there is little noise and weaken the contour enhancement when there is much noise. When the noise is particularly large, it is more effective in terms of image quality to reverse the edge enhancement to reduce the high frequency components. However, in the conventional example, careful consideration cannot be given to such noise.

[0008]

Therefore, it has a noise measuring function rather than the fixed or manual setting of the edge enhancement processing as in the conventional example, and the edge enhancement suitable for the measured noise amount and the noise amount can be adjusted according to the noise amount. The point is what kind of measures should be taken to ensure that the user always sees an image of the optimum image quality by performing the image quality setting process.

[0009]

Therefore, the present invention solves the above-mentioned problems by the structure of the following contour emphasizing device. As configuration 1, subtraction processing is performed between the same signal waveforms that are periodically arranged in the composite video signal to obtain a difference signal, and the additional noise component of the transmission system that appears in the signal period when the original value should be zero is measured. And a contour adjusting circuit that adjusts the degree of contour enhancement of the signal corresponding to the composite video signal based on the output of the circuit. When the amount is small, the degree of contour enhancement is strengthened, and when the amount of measurement noise is large, the degree of contour enhancement is weakened to carry out contour enhancement. In the configuration 3, in configuration 1, the measurement noise amount is small. In the case, the degree of contour enhancement is strengthened, the degree of contour enhancement is weakened as the measurement noise amount increases, and when the measurement noise amount further increases, the slope of the contour portion is made gentle to enhance the contour enhancement. The Migihitsuji, as a 4, Configuration 1
In the configuration 5, in the configuration 5, the signal waveform in the vertical blanking period is used as the same signal waveform periodically arranged in the composite video signal. As the same signal waveform,
Using the signal waveform of the equalization pulse period and the vertical synchronization signal period,
The difference is obtained between the fields to obtain a differential signal, and the noise component is measured. As the configuration 6, in the configuration 1, the same signal waveform periodically arranged in the composite video signal is used as the VITS period. A GCR signal waveform is used to obtain a differential signal between four fields and a noise component is measured. As Configuration 7, in Configurations 1 to 6, the same cyclically arranged in the composite video signal is used. Subtraction processing is performed between the signal waveforms of to obtain the differential signal, and the additional noise component existing in the high frequency signal region of the video signal band is measured by the high-pass filter, and the contour enhancement based on the measured noise amount of the additional noise component is performed. I try to adjust the degree.

[0010]

[Embodiment] An embodiment of the contour emphasizing device of the present invention will be described.
It will be sequentially described below with reference to the drawings. 1 to 5 show an embodiment of the present invention, and FIGS. 7 to 10 are operation explanatory diagrams thereof. In explaining the operation, a simplified simulated expression method is also used for convenience. For convenience of explanation, a signal delay due to the processing time of each circuit itself, and a delay circuit and the like used only for correcting the delay are omitted unless necessary for explanation.

FIG. 1 is an overall view of the contour emphasizing device of the present invention, FIG. 2 is a concrete configuration diagram of a noise measuring circuit 14 which is one block of FIG. 1, and FIG. 3 is a standard block which is one block of FIG. 4 is a specific configuration diagram of a deviation circuit 24. FIG. In FIG. 1, blocks 11 and 12 are blocks 6 of FIG. 6 showing a conventional example.
The contour enhancement circuit unit has the same function as 1 to 63. The block 12 combines the functions of the blocks 62 and 63 of the conventional example, and has a function of combining the functions of a high-pass filter and an amplifier. The newly added block 14 is a noise measuring circuit that controls the amplification factor of the amplifier of the block 12 based on the noise amount.

As the input 1 applied to the line La in FIG. 1, it is assumed that a luminance signal is handled and correction such as sharpening of the waveform slope portion of this signal is performed. This input signal (input 1) is connected to one input terminal of the adder (ADD) of the block 11 and the high-pass filter / amplifier (contour adjusting circuit) 12
Added to. The high-pass filter / amplifier 12 has a function of extracting horizontal and vertical two-dimensional high-pass components. Figure 4
Is a specific example of the high-pass filter, which is a transversal filter including a delay circuit group, an amplifier group (weighting circuit group), and an amplifier having a gain control function.

Although this circuit can be realized by analog, it is generally constituted by digital. La-a1 is an input line and La-a2 is an output line. Block 41A ~
41H is a group of delay circuits connected in series, each having a delay time Td (see the following equation). Further, 4fsc (fsc is a color subcarrier frequency) is a sampling frequency normally used when digitizing a TV signal.

[0014]

[Equation 1]

Nine signals with Td intervals are output before and after the delay circuit group. These delayed signal groups are supplied to the next-stage amplifier groups 42A to 42I, respectively. In this amplifier group, the input signals are each weighted by a predetermined magnification. The nine outputs from this amplifier group are added and synthesized by the next adder 43. Input line La-a1
9 to the adder 43, as shown in FIG.
= Μm (upper frequency of signal band, see Eq. 2) has a peak in the horizontal direction.

[0016]

[Equation 2]

The signal from the adder 43 is fed to the next amplifier 44.
, Where it is amplified by a factor Kh. This amplification factor Kh is transferred to the noise measuring circuit 14 via the line Ld.
As described later, the degree of edge enhancement in the horizontal direction is controlled according to the amount of noise included in the signal so that optimum image quality can be obtained.

The horizontal high-pass filtered output from the block 44 in FIG. 4 is supplied to the circuits (blocks 45A, 45B and 46) in the subsequent stages.
A to 46C, 47, 48). The high frequency range of the vertical high-pass filter here means a high-frequency region when viewed in a field image composed of 262.5 scanning lines of the interlace system, and is a frame image composed of 525 scanning lines. When viewed, it is in the middle range (meaning the middle range of the frequency range that the image can take).

The blocks 45A and 45B are delay circuits connected in series, respectively. The delay time Tv of each delay circuit is
It is set to the value of the horizontal scanning period as in the following formula. These delay circuits can be composed of line memories such as FIFO.

[0020]

[Equation 3]

The three signals having the Tv intervals output from the two delay circuits are supplied to the next amplifier groups 46A to 46C, given a predetermined weighting (amplification factor), and added by the adder 47 in the next stage. It is additively synthesized. The signal from the block 47 is fed to the next amplifier 48, where the scaling factor K
It is amplified by v. This amplification factor Kv is sent as a control signal from the noise measuring circuit 14 through the line Ld, and vertical edge enhancement is performed so as to obtain an optimum image quality in accordance with the detected noise amount as described later. Control the degree of. The output from the amplifier 48 is a line La-a2 as a two-dimensional edge enhancement component that works in the horizontal and vertical directions.
Is output via.

Next, blocks 45A, 45B and 46A-
An example of the characteristics of the circuit formed by 46C and 47C is shown in FIG.
It shows in (e). This characteristic diagram shows the characteristics of the high-pass filter in the vertical direction, which has a peak at ν = ν m / 2 (see equation 4).

[0023]

[Equation 4]

Since the TV scanning is the interlace system, the vertical high-pass filter using the line memory has a peak frequency of ν m / 2. The output of FIG. 4, that is, the output signal of the high-pass filter / amplifier 12 of FIG.
Of the input signal from the line La and added to the input signal from the line La, and output from the line Lb (output 1). As described above, the signal passing through the block 12 is a high frequency component of the input luminance signal, and the output 1 is a signal in which the high frequency region is emphasized two-dimensionally in the horizontal and vertical directions and the contour component is emphasized.

Next, the noise measuring circuit 14, which is a main part of the contour emphasizing circuit of the present invention, will be described below with reference to the drawings.
The specific circuit configuration of the noise measuring circuit 14 is shown in FIG. The line Lc in the figure shows the second noise detection target signal
The input frequency signal (input 2) of which the contained frequency component is fm
A composite video signal up to (about 4 MHz) is supplied. In this composite video signal, there is a signal portion having the same waveform in a predetermined cycle.

One example thereof is the signal portion of three lines before and after the vertical synchronizing signal in the vertical blanking period. As shown in FIG. 7, FIG. 7A shows a first field (even field) and FIG. 7B shows a second field (odd field) vertical sync signal period. It is a waveform diagram of the equalization pulse period for 3H (= 3 lines) before and after that, and these are completely the same waveforms. Therefore, one field (262.5
H) If the difference is taken between the separated signals, the result of FIG.
(A) -FIG. 7 (b)} or FIG. 7 (d) {FIG. 7 (b)
As shown in FIG. 7A, the section Tn1 from t = ta1 to tb1 is
The original signal components including the ghost are canceled out, resulting in a no-signal period.

If there is noise in this no-signal period, the noise is not canceled by subtraction because of its decorrelation,
FIG. 7 (e) {noise superimposed on FIG. 7 (c)} or FIG.
As shown in (f) {superimposed noise in FIG. 7 (d)}, the superimposed noise component is detected. When this noise is random noise such as white noise, it has the same property as the noise mixed in the composite video signal. Strictly speaking, although the characteristics are different by the comb filter formed by the difference processing by the subtraction processing for the above noise detection, they may be considered to have almost the same characteristics.

Therefore, noise can be detected in the section Tn1 (noise detectable range), but the time interval required for actually measuring the noise amount is sufficient even in a shorter period. For example, as shown in the lower part of FIG. 7, 1H period of t = tc1 to td1 in the equalizing pulse period before the vertical synchronizing signal,
That is, the range of the section Td1 can be used for noise detection. When the sampling frequency is fs (= 4fsc), about 900 sampled values are obtained, which is a sufficient number of data for measuring the noise amount.

Noise Measuring Circuit In FIG. 2, the composite video signal, which is the noise detection target signal added to the line Lc, is supplied to the subtractor of the block 21 and the delay circuit of the block 22. The delay circuit 22 receives the input signal by one signal.
Circuit (= 262.5H). In the next subtractor 21, a signal of the difference between the input from the line La and the signal one field before is formed.

When FIG. 7A shows a signal (input 2) applied to the line Lc, the signal shown in FIG. 7B is subtracted and an output as shown in FIG. 7C is obtained. When the signal of the input 2 is shown in FIG. 7B, the signal of FIG. 7A is subtracted and the output as shown in FIG. 7D is obtained. The output signal of this subtractor becomes either of FIG. 7 (c) or (d). When there is no noise in this way, the period Tn1 in the range of t = ta1 to tb1 is a period of no signal. Even if there is a ghost or the like, there is a correlation between the fields as in the composite video signal, so that the fields are canceled and disappear. However, when there is noise added in the transmission system, there is no correlation between the fields, so that it is output as a significant value.

FIG. 7E is the same as FIG. 7C when there is noise.
7F is a waveform diagram corresponding to FIG. 7F, and FIG. 7F is a waveform diagram corresponding to FIG. 7D when there is noise. Since this noise detectable range Tn1 is repeated for each field, it is possible to detect the noise state every 1/60 seconds at the shortest. However, in the noise state, if the CH of the TV is the same, it is usually that abrupt changes do not occur, so that usually one field measurement can provide sufficient accuracy.

A second example of the same waveform in a predetermined cycle in the composite video signal is VI in the vertical blanking period.
Although it is a TS (vertical interval test signal) period, it is a signal period normally used by a broadcasting station to multiplex a test signal and to make the characteristics between the stations uniform. 8 (a)-
In FIG. 8 (h), the even field is 17H.
And 18H, and an odd field is 280H and 281H.

That is, FIG. 8 (a) ... 17H and 18H of the first field
Waveform of FIG. 8 (b) ... 280H and 281H of the second field
Waveform of FIG. 8 (c) ... 17H and 18H of the third field
Waveform of FIG. 8 (d) ... 280H and 281H of the fourth field
Waveform of FIG. 8 (e) ... 17H and 18H of the fifth field
Waveform of FIG. 8 (f) ... 280H and 281H of the sixth field
Waveform of FIG. 8 (g) ... 17H and 18H of the seventh field
Waveform of FIG. 8 (h) ... 280H and 281H of the eighth field
However, nH means the n-th line.

The waveforms of 18H and 281H are reference waveforms for ghost cancellation, and are a combination waveform of a GCR waveform and an pedestal waveform of 0 field having one cycle. The waveforms of 17H and 280H are fixed waveforms in the even field and the odd field, respectively. In the figure, sig_A and si
g_B is superimposed. In the Tokyo area, NHK: 1CH, 3CH sig_A = NTC 7 composite sig_B = NTC 7 combination Private broadcast: 4CH, 6CH, 8CH, 10CH, 12CH sig_A = FCC composite sig_B = FCC Multi burst etc. Has been done.

Comparing the signals separated by four fields in FIG. 8, 17H (28) including the sync signal and the burst signal is compared.
0H) and 18H (281H) first half (t = ta2 to t)
It can be seen that b2) has exactly the same waveform. Therefore, if the delay time of the delay circuit of the block 22 of FIG. 2 is set to 4 fields, only the noise component having no video signal component can be detected by the subtractor 21. When FIG. 8 (a) is an input signal, FIG. 8 (a) is subtracted from FIG. 8 (e), and an output as shown in FIG. 8 (j) is obtained. When FIG. 8 (b) is an input signal, FIG. 8 (b) is subtracted from FIG. 8 (f), and an output as shown in FIG. 8 (i) is obtained. By performing the same operation between four fields thereafter, the output signal of this subtractor becomes either FIG. 8 (i) or FIG. 8 (j).

When there is no noise in this way, t = ta2
The Tn2 period in the range of to tb2 is a no signal period. Even if there is a ghost, etc., they are offset and disappear because of the correlation. However, in the presence of noise, the noise component added in the transmission system having no correlation between the four fields remains without being canceled and is detected. FIG. 8 (k) is a waveform diagram corresponding to FIG. 8 (i) when there is noise,
FIG. 8 (l) is a waveform diagram corresponding to FIG. 8 (j) when there is noise.

This noise detectable range is 17H (280
(H) is the Tn2 period from the leading edge of the horizontal synchronizing signal to the leading edge of the 18H (281H) GCR waveform. However, if only for noise detection, the Td2 (t = tc2 to td2) of FIG. The range is sufficient. Assuming that the sampling frequency fs is 4 fsc, there are about 500 sampled values in the Td1 period of about 1 / 2H, so that sufficient noise measurement accuracy can be obtained.

When this noise measurement period is used, the delay time of the delay circuit 22 in FIG. 2 must be set to 4 fields. However, in reality, the Td2 period is less than 1H, so four memory circuits of FIFO type with a capacity of less than 1K bytes that can store data in this period are connected in series and the data is updated every time this period comes. Since it is possible to use the delay circuit of, it is possible to realize with the circuit configuration equivalent to the case of FIG.

In the present invention, attention is paid to the regularity of these composite video signals, and the differentiator composed of blocks 21 and 22 in FIG. 2 has the same property as the noise mixed in the video signal independently of the video signal. The noise component of can be extracted. This extracted noise is applied to the contour enhancement processing, and the operation shown in FIG. 2 will be described using the signal shown in FIG. 7 as the noise measurement signal. The output of the subtractor 21 is added to the next block 23. The block 23 is a high-pass filter that operates in the horizontal direction, and has a characteristic of having a peak at the upper limit frequency μm of the video signal, as shown in FIG. 9C. The high-pass filter 23 removes a DC component, which may remain in a signal that cannot be removed by the difference unit in the preceding stage, such as APL fluctuation, and passes only a noise component effective for image quality control. Are doing

FIG. 5 shows a specific structural example of the high-pass filter 23. This circuit configuration is block 41A in FIG.
.About.41H, 42A to 42I, 43, and has the same characteristics as the high-pass filter operating in the horizontal direction of the block 12 in FIG. The setting method of such a high-pass filter considers the characteristic that noise has a random distribution in terms of space and time, and performs noise detection with the same characteristics as the high-pass filter used for the video signal. If
This is because it is possible to detect the same amount of noise as noise mixed in the horizontal high frequency components of the video signal used for edge enhancement.

Blocks 51A to 51H and 52A of FIG.
52I and 53 are blocks 41A to 41H and 42 in FIG.
It is made to correspond one-to-one with A to 42I and 43. As described with reference to FIG. 4, the blocks 51A to 51H have delay times T
a delay circuit group of d connected in series,
.About.52I are amplifier groups that give predetermined weightings (amplification factors) to the nine output signals of this delay circuit group, and a block 53 is an adder that adds these nine weighted signals. Therefore, from the output line Lb-a2, in the noise input from the input line Lb-a1,
A high frequency domain component as shown in FIG. 9C is extracted.

In FIG. 2, the next block 24 is a standard deviation circuit. FIG. 3 shows a specific circuit example. A block 31 is a squaring circuit (a function for squaring an input signal), a block 32 is an adder, 33 is a delay circuit which delays the input signal by one clock (= 1 / (4 fsc)) period, and 34 is an addition. It is a data conversion circuit having a function of dividing a signal input by the number of sample values used for the normalization and a function of obtaining a square root thereof. When the signal input from the line L2-b1 is y (t), the square root of the mean value of the sum of squares of the input y (t) in the Td1 section of FIG. That is, the standard deviation value σ is obtained by the following equation.

[0043]

[Equation 5]

In FIG. 2, the next block 25 is a memory circuit, which stores the standard deviation value σ obtained at the time when the necessary calculation is completed, and maintains the output level at the value σ. A latch circuit or the like is used for the storage circuit 25. As described in the description of FIG. 7, the waveform of FIG. 7C or 7D, that is, the waveform of FIG.
Since it is obtained for each field, the standard deviation value σ is also updated for each field.

In FIG. 2, the next block 26 is a data converter. FIG. 9B shows an example of the conversion characteristic. The input standard deviation value σ is plotted on the horizontal axis, and the contour emphasis sensitivity (control signal) K is plotted on the vertical axis. There are two types of edge enhancement sensitivity K, and Kh (FIG. 9) for controlling the degree of horizontal edge enhancement.
(B) characteristic of solid line) and Kv (characteristic of one-dot chain line in FIG. 9B) for controlling the degree of contour enhancement in the vertical direction. These conversion characteristics are as follows.

[0046]

[Equation 6]

These two contour emphasis sensitivities Kh and Kv are
As the output 2, through the line Ld, the high-pass filter of FIG.
It is supplied to the amplifier of the amplifier 12, and Kh is the amplifier 4 of FIG.
4 and Kv are applied as control signals to the amplifier 48 of FIG. 4, respectively, and serve to control the output magnification of the horizontal high-pass filter and the vertical high-pass filter. In FIG. 9B, when the value at the inflection point of the control signal (edge emphasis sensitivity) K is determined as in the following equation,

[0048]

[Equation 7]

The contour enhancement degree Kh acting in the horizontal direction obtains the maximum contour enhancement effect (Ko) in 0 ≦ σ <σmin, and in the range of σmin ≦ σ <σmid, as the value of σ increases, The contour enhancement effect becomes weaker, and at σ = σmid, there is no contour enhancement effect and it becomes 0. In the range of σmid ≤ σ <σmax, the contour enhancement effect becomes a negative value, and the high range of the luminance signal is attenuated. Will be done.

Further, in the range of σ max ≤σ, the maximum high frequency component attenuation effect (-Ko) is obtained. On the other hand, the edge enhancement degree K that operates in the vertical direction
v obtains the maximum contour enhancement effect (Ko) in 0 ≦ σ <σmin, and in the range of σmin ≦ σ <σmid, the contour enhancement effect becomes weaker as the value of σ increases, and σ = At σmid, the contour enhancement effect disappears and becomes zero, and even in the range of σmid <σ, the contour enhancement effect becomes zero.

The reason why the horizontal and vertical directions have different functions is that the characteristics of the high-pass filter in the vertical direction have a peak at a frequency in the middle of the signal existence region ν m. This is because the characteristics that are different from those of the horizontal high-pass filter are taken into consideration. FIG. 9D is a characteristic showing how the frequency characteristic of the signal in the horizontal direction changes depending on the control signal Kh acting in the horizontal direction. This corresponds to a horizontal impulse response. The characteristic of d1 in the figure is
A signal having the frequency characteristic as shown in FIG. 9A is added to the input 1 of FIG. 1, and the characteristic of Kh of FIG. 9B becomes (σ = σmin, K = K
It is a horizontal frequency characteristic of the output 1 at the time of (max). At this time, the circuit of FIG. 1 has a characteristic of emphasizing the high range in the horizontal direction and emphasizing the contour.

The characteristic of d2 in FIG. 9D is the same as the input 1 in FIG.
A signal having the frequency characteristic shown in FIG. 9A is added to
It is a horizontal frequency characteristic of the output 1 when the characteristic of Kh in (b) is (σ = σmid, K = Kmid). At this time, the signal of the input 1 is taken out as the output 1 as it is. The characteristic of d3 in the figure is as shown in FIG.
FIG. 9 when a signal having a frequency characteristic as shown in FIG.
The characteristic of Kh in (b) is (σ = σmax, K = Kmin = -Ko)
It is a horizontal frequency characteristic of the output 1 at the time of. At this time, the high frequency components of the signal at input 1 are attenuated, and the circuit of FIG. 1 functions as a horizontal low pass filter.

FIG. 10 (f) shows a control signal K acting in the vertical direction.
This is a characteristic showing how the frequency characteristic of the signal in the vertical direction changes depending on v, and is an impulse response in the vertical direction. In the characteristic of f1 in the figure, a signal having a flat characteristic is added to the input 1 of FIG. 1 up to the upper limit spatial frequency ν = νm (= 525 / 2cph) in the vertical direction, and the characteristic of Kv in FIG. σ
= Σmin, K = Kmax), the frequency characteristics of the output 1 in the vertical direction. At this time, the circuit of FIG. 1 has a characteristic of emphasizing the high range in the vertical direction and emphasizing the contour. The characteristic of f2 in the figure is that the upper limit spatial frequency ν = νm in the direction perpendicular to the input 1 in FIG.
Signals with flat characteristics are added up to (= 525 / 2cph).
It is a vertical frequency characteristic of the output 1 when the Kv characteristic of (b) is (σ = σmid, K = Kmid). At this time, the signal of the input 1 is taken out as the output 1 as it is.

FIG. 10 (g) is a time domain waveform diagram showing the effect of the contour correction in the horizontal direction in order to visually express the effect of the contour correction. In FIG. 10 (g), assuming that a step waveform with an upward slope such as g2 is the signal of input 1 in FIG. 1, the waveform of g1 is obtained as output 1 when Kh = Kmax (= Ko). It is a waveform diagram shown in FIG.
It is a waveform diagram obtained as an output 1 when Kmid (= 0), and a waveform diagram of g3 is a waveform diagram obtained as an output 1 when Kh = Kmin (= -Ko). Thus, FIG.
When g2 in (g) is an input waveform, g1 is a waveform with contour enhancement, g2 is a waveform without contour enhancement processing, and g3 is a waveform with contour blunting. In this way, the degree of contour enhancement can be controlled by the standard deviation σ that changes based on the amount of noise.

FIG. 10H is a diagram in which the two-dimensional spatial frequency region in which the horizontal and vertical high-pass filters function is shown by hatching. The diagonally downward-sloping area indicated by "H" is an area where the horizontal high-pass filter functions, and the upward-sloping diagonally area indicated by "V" is an area where the vertical high-pass filter operates. These shaded areas are areas within -6 dB of the peak value of the high-pass filter.

As described above, the pure noise amount having the same property as the noise component in the composite video signal is measured, and its magnitude is measured as the standard deviation value σ. Based on this σ value, when the noise is small, the waveform is calculated. It emphasizes the slanted part, that is, the contour, and weakens the contour enhancement when there is a lot of noise, or performs negative contour enhancement to lower the high range, but at the cost of lowering the high range of the signal, but it does not cause a rough pattern due to noise. By setting, it is possible to obtain visually good image quality.

The vertical blanking period as shown in FIG. 7 is also present in TV systems such as PAL and SECAM in Europe, and although the synchronization cycle is slightly different from that in the NTSC system, it is different for each field. Since the same waveform is repeated, the present invention can be applied for the same purpose.

Further, it is of course possible to use the same signal waveform for a period having a horizontal synchronizing signal and a burst signal, which are separated by two lines (H). In this case, the delay time of the delay circuit may be short, which simplifies the circuit. Although the standard deviation value is used when evaluating the amount of noise to be measured, the average value of the sum of squares, that is, the variance value can be used, or the average value of the sum of absolute values can be used. . However, in this case, it is necessary to set conversion characteristics to Kh and Kv that are suitable for each value.

Further, in measuring the noise amount, the noise itself can be measured without being influenced by the ghost, and therefore, in the edge enhancement processing using this, correction corresponding to the noise amount mixed in the signal is performed. become. Although the luminance signal has been taken as an example and described as the edge enhancement signal, various signals such as a color signal, a color difference signal, and an RGB signal can be selected as a target. However, the signal to be subjected to the noise measurement is a composite video signal which includes information of the transmission system and in which the same waveform having a constant cycle remains unprocessed.

[0060]

As described above, the contour emphasizing device of the present invention has the following effects. (B) The subtraction process is performed between the same signal waveforms periodically arranged in the composite video signal to obtain a difference signal, and the additional noise component of the transmission system that appears in the signal period when the original value should be zero is detected. The noise having the same property as the noise mixed in the video signal can be measured without being influenced by the contents of the video signal, and thus objective and accurate contour enhancement can be performed. (B) If there is a lot of noise, reverse the effect of normal contour enhancement,
In other words, it can be used as a negative value to reduce high frequency components and smooth the rough noise screen.
It enables a wider range of operations than the conventional contour enhancement function. (C) When the vertical blanking period is used, the measurement noise amount does not include the ghost of the video signal and the synchronization signal, and the image quality can be improved with high accuracy. (D) The conventional image quality adjustment is a manual type, but it is possible to perform automatic image quality adjustment without bothering the user. (E) Not limited to television receivers, it can be applied not only to VTRs, but also to European TVs such as PAL and SECAM.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a contour enhancement device of the present invention.

FIG. 2 is a concrete configuration diagram of a noise measuring circuit of an embodiment of the contour enhancing apparatus of the present invention.

FIG. 3 is a specific configuration diagram of an embodiment of a standard deviation circuit of the present invention.

FIG. 4 is a specific configuration diagram of an embodiment of a high frequency filter according to the present invention.

FIG. 5 is a specific configuration diagram of an embodiment of a high-pass filter according to the present invention.

FIG. 6 is a block configuration diagram of a conventional contour enhancement apparatus.

FIG. 7 is an operation explanatory diagram of the contour enhancement apparatus of the present invention.

FIG. 8 is an operation explanatory diagram of the present invention.

FIG. 9 is an operation explanatory diagram of the present invention.

FIG. 10 is an operation explanatory diagram of the present invention.

[Explanation of symbols]

11, 32, 43, 47, 53 Adder (ADD) 12 High-pass filter / amplifier (contour adjusting circuit) 23 High-pass filter 14 Noise measurement circuit 21 Subtractor (SUB) 22, 33, 41A to 41H, 45A , 45B, 51A
-51H Delay circuit 24 Standard deviation circuit 25 Storage circuit 26,34 Data converter 31 Square circuit 42A-42I, 44, 46A-46C, 48, 52A
-52I Amplifier K Contour emphasis sensitivity Kh Horizontal contour emphasis sensitivity Kv Vertical contour emphasis sensitivity Td1, Td2 Noise detection range Tn1, Tn2 Noise detectable range σ Standard deviation value

Claims (7)

[Claims] 1. A differential signal is obtained by performing a subtraction process between identical signal waveforms which are periodically arranged in a composite video signal, and an additional noise component of a transmission system which appears in a signal period when the original value should be zero is measured. And a contour adjusting circuit that adjusts the degree of contour enhancement of the signal corresponding to the composite video signal based on the output of the noise measuring circuit. . 2. The contour enhancement according to claim 1, wherein the degree of contour enhancement is increased when the measurement noise amount is small, and the degree of contour enhancement is weakened when the measurement noise amount is large. Contour enhancement device. 3. The method according to claim 1, wherein when the amount of measurement noise is small, the degree of contour enhancement is increased, and when the amount of measurement noise is increased, the degree of edge enhancement is weakened and the amount of measurement noise is further increased. Is a contour emphasizing device characterized by grading the inclination of the contour portion to perform contour emphasizing. 4. The contour enhancing apparatus according to claim 1, wherein a signal waveform in a vertical blanking period is used as the same signal waveform periodically arranged in the composite video signal. 5. A signal waveform of an equalizing pulse period or a vertical synchronizing signal period is used as the same signal waveform periodically arranged in a composite video signal according to claim 4, to obtain a difference between fields. A contour enhancement device characterized by obtaining a differential signal and measuring a noise component. 6. The noise according to claim 1, wherein a GCR signal waveform in the VITS period is used as the same signal waveform periodically arranged in the composite video signal to obtain a differential signal between four fields. A contour enhancement device characterized by measuring a component. 7. The high-pass filter according to claim 1, wherein subtraction processing is performed between the same signal waveforms periodically arranged in the composite video signal to obtain a differential signal, and the high-pass filter further increases the high frequency of the video signal band. An outline emphasis apparatus, characterized in that it measures the additional noise component existing in a signal region and adjusts the degree of outline emphasis based on the amount of measurement noise of the additional noise component.