JPH0410794A - Picture quality correcting circuit - Google Patents

Abstract

PURPOSE:To improve the picture quality of an image with low brightness by outputting a level control signal proportional to the average level of a brightness signal from a level detecting circuit and allowing a brightness level converting circuit to emphasize the low level part of the brightness signal when the level control signal is low, and execute linear processing at the time of a high level control signal. CONSTITUTION:The picture quality correcting circuit is provided with the level detecting circuit 7 for inputting a brightness signal included in a TV signal, detecting the average level of the brightness signal and outputting a level control signal and the brightness level converting circuit 8 for inputting the brightness signal, selecting I/O characteristics in accordance with the level control signal outputted from the circuit 7, converting the level of the brightness signal, and outputting the converted signal. When the average level of the brightness signal is low, the circuit 8 selects non-linear I/O characteristics increasing the emphasis value of the low level part of the brightness signal, and at the time of a high average value, selects I/O characteristics close to linear shape. Thereby, the low level part of the image with low brightness can be cleared without white saturation of the high level part of the brightness signal.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to video signal display devices such as color television receivers and displays, and video signal processing devices such as VTRs, and particularly to digital signal processing for component television signals. The present invention relates to an image quality correction circuit that can improve the image quality of low-luminance images in a digital signal processing device that performs.

[Conventional technology]

Conventionally, methods for improving the image quality of low-luminance images include, for example,
There is a method described below that controls the color level according to the average level of the luminance signal and reduces the influence of color noise in low-luminance images.

In the conventional technology shown in "Color Level Control Device" described in Japanese Utility Model Application Publication No. 1162983, an integrating circuit for integrating a luminance signal and a means for converting a color difference signal level are installed, and the luminance is adjusted according to the output of the integrating circuit. Average I of the signal
When the average level of the luminance signal is high, the color difference signal level is uniformly raised, and when the average level of the luminance signal is low, the color difference signal level is uniformly lowered. This compensates for the lack of color difference signal level and reduces the influence of color noise when the average level of the luminance signal is low.

[Problem to be solved by the invention]

However, the image quality correction according to the conventional technique described above is insufficient to improve the image quality of low-luminance images, and there are problems as described below.

(1) In general, in low-luminance images, most of the luminance signals are concentrated in the low-level area, so there is a problem that the dynamic range cannot be fully utilized, and
Changes in the low-level portion of the luminance signal are difficult for the human eye to discern, so there is a problem that images are extremely difficult to see. However, the conventional technology described above is difficult to distinguish between such luminance signal levels. was not given any consideration.

If we try to apply the above-mentioned conventional technology to solve such a problem, for example, we can apply linear processing to uniformly increase the luminance signal level in a low-luminance image (i.e., when the average level of the luminance signal is low). If the brightness signal is supposed to be concentrated in the low level part, but some high level part is included, the high level part will exceed the guinamin range, resulting in crushed whites. The problem is that it can cause

Therefore, nonlinear processing is required to emphasize only the low-level portion of the luminance signal in a low-luminance image.

(2) Additionally, in general, changes in the low level part of color difference signals are easily detected by the human eye as large changes in hue;
It is well known that in low-brightness images, the human eye can easily detect noise in high-level parts of color difference signals. Since linear processing is applied to uniformly lower the color difference signal level at low levels), noise in the high level part of the color difference signal can be reduced to some extent, but the change in the low level part of the color difference signal also increases. There was a problem in that the change in hue was noticeable.

For this reason, nonlinear processing is required to reduce only the high-level portion of the color difference signal in a low-luminance image.

(3) Furthermore, in the above conventional technology, the average level of the luminance signal is detected using an integrating circuit, but if the integration period of this integrating circuit is too short, the color difference signal If the integration period is too long, the color difference signal level will not be able to follow the changes in the brightness signal, even if the noise and brightness changes are small. For example, there is a problem in that the image quality deteriorates when there is a scene change in a still image, and it is therefore difficult to set the integration period.

The present invention has been made in view of the above-mentioned problems of the prior art, and therefore, a first object of the present invention is to prevent whitening of the high level portion of the luminance signal in a low luminance image. An object of the present invention is to provide an image quality correction circuit that can clarify low-level parts. A second object of the present invention is to provide an image quality correction circuit that can reduce noise in the high level part of the color difference signal in a low brightness image without changing the low level part of the color difference signal that is easily detected as a change in hue. It is about providing. Furthermore, the third object of the present invention is to
When there is a lot of noise or brightness changes, the image quality will not deteriorate due to frequent changes in the brightness signal or color difference signal level.
Another object of the present invention is to provide an image quality correction circuit that can quickly follow changes in brightness signals and perform image quality correction when noise and changes in brightness are small.

[Means to solve the problem]

In order to achieve the first object among the above-mentioned objects, the present invention inputs a luminance signal included in a component television signal to an image quality correction circuit, detects the average level of the luminance signal, and detects the average level of the luminance signal. a level detection circuit that outputs a detection result as a level control signal; and a level detection circuit that receives the luminance signal;
a brightness level conversion circuit that selects an input/output characteristic according to the level control signal from the level detection circuit, converts and outputs the level of the brightness signal according to the selected input/output characteristic; A color difference level conversion circuit that manually inputs the included color difference signal, selects an input/output characteristic according to the level control signal from the level detection circuit, and converts and outputs the level of the color difference signal according to the selected human output characteristic. The brightness level conversion circuit is configured to:
When the average level of the luminance signal is low, a nonlinear input/output characteristic is selected in which the amount of emphasis on the low-level portion of the luminance signal is increased, and when the average level of the luminance signal is high, the human output is closer to linear. Now you can select characteristics.

Further, in order to achieve the second object, in the present invention, the color difference level conversion circuit may, in accordance with the level control signal,
The lower the average level of the luminance signal is, the more non-linear input/output characteristics are selected in which the amount of de-emphasis of the higher level portion of the color difference signal is increased, and the higher the average level of the luminance signal is, the more linear the input/output characteristic is selected. Output characteristics can now be selected.

Furthermore, in order to achieve the third object, the present invention provides the level detection circuit as a one-frame level detection circuit that inputs the luminance signal, detects and outputs the average level of the luminance signal for each screen. and a differentiator which inputs the output signal from the one frame level detection circuit, detects and outputs the rate of change of the output signal, and inputs the output signal from the one frame level detection circuit and outputs the output signal. an integrator that integrates and outputs the level control signal as the level control signal;
The integration period is made longer as the rate of change of the output signal from the frame level detection circuit is higher.

[Operation] In the above means, the level detection circuit outputs a level control signal proportional to the mean level of the luminance signal, and when the luminance 1
/Hel conversion circuit is I li when the level control signal is low.
(The low level part of the g signal is emphasized, and conversely, when the level control signal is high, it is not emphasized and linear processing is performed.

As a result, the luminance signal is non-linearly processed based on the difference in the average level of the luminance signal, and in a low-luminance image, the low-level portion of the luminance signal can be made clear without causing whiteout in the high-level portion of the luminance signal.

Moreover, since the nonlinear processing is performed by a digital circuit, variations in characteristics can be suppressed.

In addition, in the means described in item 1, the color difference level conversion circuit always linearly processes the low level part of the color difference signal, deemphasizes the high level part when the level control signal is low, and does not deemphasize the high level part when the level control signal is high. Process.

As a result, the color difference signal is processed non-linearly based on the difference in the average level of the luminance signal, and noise in the high level part of the color difference signal can be reduced in low brightness images without changing the low level part of the color difference signal that is easily detected as a hue change. can be reduced.

Further, in the above means, the one frame level detection circuit in the brightness level detection circuit outputs a signal proportional to the average level of the brightness signal for each screen, and the integrator integrates the output of the one frame level detection circuit for several screens. to generate a level control signal. Further, the differentiator sets the integration period of the integrator in proportion to the rate of change in the output of the one-frame level detection circuit. As a result, when there is a lot of noise or brightness changes, the integration period becomes long, and image quality deterioration such as flicker due to frequent changes in the level of the brightness signal or color difference signal does not occur. Furthermore, when there is little noise or brightness change, the integration period becomes short, and it is possible to quickly follow changes in the brightness signal and perform image quality correction.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG.

FIG. 1 is a block diagram showing an embodiment of the present invention, in which the image quality correction circuit of this embodiment is installed in a television receiver that digitally processes a composite I television signal, converts it into a component television signal, and outputs it. This shows the case where .

In Figure 1, 1 is an input terminal for inputting a composite television signal, and 2 is an analog/digital (hereinafter referred to as
3 is a YC separation circuit that separates a luminance signal (hereinafter abbreviated as Y) and a color difference signal (hereinafter abbreviated as C) from a composite television signal into components; 4 is a Y processing circuit; 5 is a C processing circuit, the part surrounded by the dotted line 6 is an image quality correction circuit of this embodiment, 7 is a level detection circuit, 8 is a Y level conversion circuit, 9 is a C level conversion circuit, 10
1 is a synchronizing signal generating circuit, 1.1 and 12.13 are digital/analog (hereinafter abbreviated as D/A) converters, and A is a level control signal.

First, a composite television signal inputted from an input terminal 1 is converted into a digital signal by an A/D converter 2 and supplied to a YC separation circuit 3 and a synchronization signal generation circuit 10. The composite television signal is Y in YC separation circuit 3.
On the other hand, the synchronizing signal generating circuit [0] generates control signals necessary for processing within the same television receiver in synchronization with the composite television signal.

Y and C output from the YC separation circuit 3 are supplied to a Y processing circuit 4 and a C processing circuit 5, respectively.Y is decoded in the Y processing circuit 4, and C is decoded in the C processing circuit 5. Color difference signal (hereinafter referred to as C, C2. For example, C
,,C2 are ■ and Q in the NTSC system. ). Y, C, . . . C2 converted into component television signals through the above processing are supplied to the image quality correction circuit 6, and subjected to the image quality correction processing described below.

First, Y is supplied to the level detection circuit 7 and the Y level conversion circuit 8, and C, . . . C2 are supplied to the C level conversion circuit 6. In the level detection circuit 7, Y is compared with a preset reference value and converted into a signal indicating the level of Y level. Furthermore, the signal indicating the level of the Y level is integrated and converted into a level control signal A proportional to the average level of Y, which is output to the Y level conversion circuit 8 and the C level conversion circuit 9.

On the other hand, in the Y level conversion circuit 8, a plurality of input/output characteristics for converting the Y level are set in advance, and one input/output characteristic is selected from these according to the level control signal A.
The input Y level is converted based on this input/output characteristic. Similarly 1. The C level conversion circuit 9 also has a plurality of input/output characteristics set in advance for converting the C level.
One input/output characteristic is selected from among these according to the level control signal A, and the levels of C, C2 are similarly converted by two circuits based on this input/output characteristic.

Y, Ci whose image quality has been corrected by the above level conversion processing
, Cz are converted back to analog signals by D/A converters 11, 12, and 13, respectively, and output to, for example, a display.

Next, the operation of each part of this embodiment is shown in FIGS. 2, 3, 4,
This will be explained in detail using FIG.

FIG. 2 is a characteristic diagram showing an example of the input/output characteristics of the Y level conversion circuit 8 of FIG. 1, and shows the case where the level control signal A is set to three levels.

In Figure 2, the horizontal axis is the input level of Y, the vertical axis is the output level of Y, a is the input/output characteristic when the level control signal A is low (when the average level of Y is low), and b is the level control signal A. C is the human output characteristic when is medium (when the average level of Y is medium), and C is the input/output characteristic when the level control signal A is high (when the average level of Y is high).

First, when the level control signal A is low, the Y input to the Y level conversion circuit 8 is level converted and outputted according to the input/output characteristic a in which the low level portion is emphasized. Furthermore, when the level control signal A is medium, the signal is level-converted and output using an input/output characteristic in which the amount of emphasis in the low level portion is reduced compared to the input/output characteristic a.

Furthermore, when the level control signal A is high, the signal is level-converted and outputted using a conversion characteristic C that is close to a linear characteristic, regardless of the input level.

In this manner, the Y level conversion circuit 8 can perform nonlinear processing on the low level portion of Y in accordance with the level control signal A, such that the amount of emphasis increases as the average level of Y decreases. As a result, the image quality can be made clearer without causing whiteout in the high-level portion of Y in a low-luminance image.

Here, the Y level conversion circuit 8 is configured with a memory such as a ROM, and can be easily realized by using a table lookup method. Moreover, just like that, RO
When configured with memories such as M, it is possible to suppress variations in characteristics that tend to occur when nonlinear processing is performed.

In addition, in the explanation using FIG. 2 above, the level control signal A is
Although this is explained as a level, in the present invention, at least 2
It is fine as long as it is above the level. Note that the number of input/output characteristics in the Y level conversion circuit 8 at that time is the number of levels of the level control signal A minus G.

Next, FIG. 3 is a characteristic diagram showing an example of input/output characteristics in the C level conversion circuit 9 of FIG.
The figure shows the case where 3 levels are set.

In Figure 3, the horizontal axis is the input level of C7 and C2, and the vertical axis is the input level of C7 and C2.
Qj is the output level of C1 and C2, d is the level control signal A
Input/output characteristics when is low (when the average level of Y is low), e
is the input/output characteristic when the level control signal A is medium (when the average level of Y is medium), and f is the input/output characteristic when the level control signal A is high (when the average level of Y is high) .

First, when the level control signal A is low, C, . Furthermore, when the level control signal A is medium, the signal is level-converted and output using the input/output characteristic e, which has a reduced amount of de-emphasis in the high level section compared to the input/output characteristic d. Furthermore, when the level control signal A is high, the signal is level-converted and output using a conversion characteristic f that is almost linear, regardless of the human power level.

In this way, the C level conversion circuit 9 always performs linear processing on the low level parts of CI and C2 according to the level control signal A, and applies the average level of Y to the high level parts of C, C2. It is possible to perform non-linear processing in which the amount of de-emphasis increases the lower the value is. As a result, noise in the high level portion of the color difference signal can be reduced in a low brightness image without changing the low level portion of the color difference signal that is easily detected as a hue change.

In addition, in the explanation using FIG. 3 above, the level control signal A is
Although the explanation is given in terms of levels, in the present invention, it is sufficient that there are at least two or more levels. Note that the number of input/output characteristics in the C level conversion circuit 9 at that time matches the number of levels to the level control signal.

Furthermore, the C level conversion circuit 9 converts the levels by providing two circuits for the two inputs C, C2.
In the present invention, it is sufficient to convert the levels of CI and C2 in the same way. For example, if they can be treated as one system of signals by orthogonal multiplexing or line sequential multiplexing, the C level conversion circuit 9 can be constructed of one system of circuits.

Next, FIG. 4 is a block diagram showing a specific example of the level detection circuit 7 in FIG. 1, in which the average L/hair 1/
A case is shown in which the level control signal A is divided into low, middle, and high levels and three levels of level control signal A are generated.

In Figure 4, the part surrounded by the dotted line of 401 is 701
This is a one-frame level detection circuit that detects the average level for each screen, and for example, the first clock generation circuit 402 and
It is composed of a first up/down counter 403 and a one-screen level detection circuit 413. Furthermore, the part surrounded by the dotted line 404 is an integrator that integrates the output of the one-frame level detection circuit 401, and for example, it includes a second clock generation circuit 405, a second up/down counter 406, and a level detection circuit. 414. Further, the part surrounded by the dotted line 407 is a differentiator that detects the rate of change in the output of the one-frame level detection circuit 401. For example, the part surrounded by the data latch 408, the subtracter 409, and the third clock generation circuit 410 , a 30th counter 411, and a reset pulse generation circuit 412.

Here, the level control signal A output from the integrator 404 corresponds to that shown in FIG. It is output from the synchronization signal generation circuit 10. In addition, the input signal Y is input to the Y processing circuit 4 in FIG.
matches the output of

First, Y output from the Y processing circuit 4 in FIG. 1 is supplied to a one-frame level detection circuit 401, where it is converted into a signal proportional to the average level of Y for each frame.

The operation of this one frame level detection circuit 401 will now be explained.

In the l-frame level detection circuit 401, Y is first supplied to the first clock generation circuit 402, where it is compared with a preset reference value, and for example, the Y level is determined as low, medium, high, etc. is converted into a signal indicating

Further, a first clock for operating the first up/down counter 403 is generated by signals indicating these three levels. Next, this first clock is supplied to the first up/down counter 403, and Y
When the level is low, the count value is decreased, when the level is medium, the count value is not changed, and when the level is high, the count value is increased. In this way, the first up/down counter 403 counts according to the first clock and is reset every screen period by the frame pulse. Next, the one-screen level detection circuit 413 compares the count value of the first up-down counter 403 immediately before resetting with a preset reference value, and generates a signal indicating high, medium, or low level depending on the count value. converted.

The above is the operation of the one-frame level detection circuit, and the output of the one-screen level detection circuit 413 is output to the integrator 404 and the differentiator 407 as a signal indicating the average level of Y for one screen.

Next, in the differentiator 407, the one frame level detection circuit 40
1 is detected, and the integration period of the integrator 404 is set according to this change rate.

Now, the operation of the differentiator 407 will be explained below.

In the differential 1407, first, the output of the one-frame level detection circuit 401 is supplied to the data latch 408 and the subtracter 409, where it is subtracted from the average level of Y of one screen before, and the difference signal of this subtraction result is used as the third The signal is supplied to clock generation circuit 410. Next, in the third clock generation circuit 410,
The difference signal generates a third clock that operates the third counter 411. Next, this third clock is supplied to the third counter 411, and for example, if a difference occurs, the count value is increased, and if there is no difference, the count value is not changed.

In this manner, the third counter 411 counts for a certain period of time according to the third clock, and this count value is output to the reset pulse generation circuit 412 as a signal indicating the rate of change in the average level of Y.

Next, the reset pulse generation circuit 412 generates a reset pulse whose period becomes longer when the rate of change of the average level of Y is high, based on the output of the third counter 411.

The above is the operation of the differentiator 407, and the reset pulse from the reset pulse generation circuit 412 is applied to the integrator 404.
It is output as a reset pulse that determines the integration period.

On the other hand, in the integrator 404, the one frame level detection circuit 4
The output of 01 is integrated with the period of the reset pulse output from the differentiator 407 to generate the level control signal A.

Now, the operation of the integrator 404 will be explained below.

In the integrator 405, first, the output of the one frame level detection circuit 401 is supplied to the second clock generation circuit 405, which operates the second up/down counter 406 similarly to the first clock generation circuit 4.2. A second clock is generated to cause the clock to run.

Next, the second up/down counter 406 counts according to the second clock and is reset by a reset pulse output from the differentiator 407. Furthermore, the level detection circuit 414 compares the count value of the second up/down counter 406 immediately before resetting with a preset reference value, and converts it into a signal indicating low, medium, or high level.

The above is the operation of the integrator 404, and the level detection circuit 41
4 is output as level control signal A.

According to this specific example, the level detection circuit 7 outputs a signal proportional to the average level of each screen of Y from the one frame level detection circuit 401, and the integrator 404 converts the output of the one frame level detection circuit 401 into a number. A level control signal A is generated by integrating over 100 planes. Further, the differentiator 407 is connected to the integrator 40 in proportion to the rate of change of the output of the one frame level detection circuit 401.
Set the integration period of 4. As a result, the integration period becomes longer when there is a lot of noise or brightness change, so that image quality deterioration such as flicker due to frequent level changes of the brightness signal or color difference signal does not occur. Furthermore, since the integration period becomes short when there is little noise or brightness change, it is possible to quickly follow changes in the brightness signal and perform image quality correction.

In addition, in the explanation in FIG. 4 above, the level control signal A
Although the explanation has been made assuming that there are three levels, in the present invention, it is sufficient that there are at least two or more levels.

Next, another specific example of the level detection circuit 7 shown in FIG. 1 will be explained using FIG. 5.

FIG. 5 is a block diagram showing another specific example of the level detection circuit in FIG. 1.

In FIG. 5, 501 is the first
This is a low pass filter (hereinafter abbreviated as LPF). Further, the part surrounded by the dotted line 401 is a one frame level detection circuit, for example, the first clock generation circuit 402
, a first up/down counter 403 , and a one-screen level detection circuit 413 . Further, 502 is a second L that integrates the output of the one frame level detection circuit 401.
It is PF.

Here, the level control signal A1 frame pulse and the input signal Y have the same meaning as explained in FIG. 4.

First, the low frequency component of the input signal Y is extracted to the output of the first LPF 501. Here, the first LPF 501 is a low-pass filter for each pixel, and for example, the first LPF
By using the output of 501, the data rate of Y can be lowered.

Next, this low frequency component of Y is detected by the one frame level detection circuit 4.
01, and is converted into a signal proportional to the average level of Y for each screen.

The operation of this one frame level detection circuit 401 will now be explained.

In the one-frame level detection circuit 401, the low frequency component of Y is first supplied to the first clock generation circuit 402. In the first clock generation circuit 402, the first LPF 501
The output is compared with a preset reference value and converted into a signal indicating the level.

Further, a first clock for operating the first up/down counter 403 is generated by a signal indicating this level. Here, by lowering the data rate of Y using the output of the first LPF 501, the frequency of the first clock can be lowered compared to the first clock in the fourth specific example described above. Next, the first up/down counter 403 counts according to this first clock, and counts up and down according to the frame pulse.
It is reset every screen cycle. Next, the one-screen level detection circuit 413 compares the count value of the first up-down counter 403 immediately before resetting with a preset reference value, and converts it into a signal indicating high, medium, or low level.

The above is the operation of the one frame level detection circuit 401,
The output of the one-screen level detection circuit 413 is output to the integrator 404 as the average level of one screen of Y.

Next, the output of the one frame level detection circuit 401 is supplied to the second LPF 502. Here, the second LPF50
2 is a low-pass filter for each frame, which integrates the output of the one-frame level detection circuit 401 and prevents false detection of the average level of Y, which occurs when there is a lot of noise or brightness changes between screens, for example. I can do it. This second
The output of the LPF 502 is output as a level control signal A.

According to this specific example, the first LPF 501 in the one-frame level detection circuit 401 lowers the data rate of Y and reduces the frequency of the clock generated by the first clock generation circuit 402. , the circuit scale of the first up/down counter 403 can be reduced. In addition, the second LPF 502 removes high-frequency components of the output of the one-frame level detection circuit 401 to prevent false detection. Circuits such as differentiators can be reduced.

〔Effect of the invention〕

The present invention has the following effects.

(1) In a low-luminance image, it is possible to brighten and thin out the low-level portions of the luminance signal without causing whiteout in the high-level portions of the luminance signal.

(2) Noise in the high level portion of the color difference signal can be reduced in a low brightness image without changing the low level portion of the color difference signal that is easily detected as a change in hue.

(3) When there is a lot of noise and brightness changes, image quality deterioration such as flicker does not occur due to frequent changes in the level of the brightness signal and color difference signal; It is possible to quickly follow changes in image quality and apply image quality corrections.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a characteristic diagram showing an example of input/output characteristics in the Y level conversion circuit 8 shown in FIG. 1, and FIG. 3 is a C level conversion circuit shown in FIG. 1. A characteristic diagram showing an example of input/output characteristics in the circuit 9, FIG. 4 is a block diagram showing one specific example of the level detection circuit 7 in FIG. 1, and FIG. 5 shows another specific example of the level detection circuit 7 in FIG. FIG. 2 is a block diagram illustrating an example. Explanation of symbols 2... A/D converter, 6... Image quality correction circuit, 7...
・Level detection circuit, 8...Y level conversion circuit, 9...
-C level) Ly conversion circuit, 401... 1 frame level detection circuit, 404... Integrator, 407... Differentiator, A... Level control signal. Agent Patent Attorney Akio Namiki

Claims (1)

[Claims] 1. In a digital signal processing device that performs digital signal processing on a digital component television signal, a luminance signal included in the component television signal is input, and an average level of the luminance signal is detected. , a level detection circuit that outputs the detection result as a level control signal; and a level detection circuit that receives the luminance signal, selects an input/output characteristic according to the level control signal from the level detection circuit, and selects an input/output characteristic according to the selected input/output characteristic. A brightness level conversion circuit that converts and outputs the level of the brightness signal and a color difference signal included in the component television signal are input, and input/output characteristics are selected according to the level control signal from the level detection circuit. , a color difference level conversion circuit that converts and outputs the level of the color difference signal according to the selected input/output characteristic, and the brightness level conversion circuit converts the level of the brightness signal according to the level control signal. When the average level is low, a non-linear input/output characteristic is selected in which an amount of emphasis on a low-level portion of the luminance signal is increased, and when the average level of the luminance signal is high, an input/output characteristic that is close to linear is selected. An image quality correction circuit characterized by: 2. In the image quality correction circuit according to claim 1, the color difference level conversion circuit de-emphasizes a portion where the level of the color difference signal is higher when the average level of the luminance signal is lower, according to the level control signal. An image quality correction circuit characterized in that a nonlinear input/output characteristic with an increased amount is selected, and the higher the average level of the luminance signal, the more linear the input/output characteristic is selected. 3. In the image quality correction circuit according to claim 1 or 2, the level detection circuit is a one-frame level detection circuit that receives the luminance signal, detects and outputs an average level of the luminance signal for each screen. , a differentiator that inputs the output signal from the 1-frame level detection circuit, detects and outputs the rate of change of the output signal; and a differentiator that inputs the output signal from the 1-frame level detection circuit and integrates the output signal. and an integrator that outputs the level control signal as the level control signal, and the integrator has a high rate of change of the output signal from the one-frame level detection circuit according to the output signal from the differentiator. An image quality correction circuit characterized by increasing its integration period as the time increases.