KR100230261B1 - Picture quality enhancing method of color signal based on the quantized mean-separate histogram equalization and the color compensation, and circuit thereof - Google Patents

Abstract

The image quality improvement method and its circuit of the present invention extracts a luminance signal from color signals, quantizes the luminance signal extracted, divides the input luminance image into a predetermined number of sub-images based on the average, Histogram equalization to output the changed luminance signal, and outputting the compensated color signals by changing the color signals to have the same ratio as the changed luminance signal, thereby providing a distortion-free color signal while improving the contrast of the color image.

Description

Method and circuit for improving image quality of color signal using quantized mean separation histogram equalization and color compensation

The present invention relates to a method and a circuit for enhancing the quality of a color signal using quantized mean separation histogram equalization and color compensation and quantizes the inputted luminance image and divides the input luminance image into a predetermined number of sub- The present invention also relates to a method and a circuit for improving the image quality by changing the color signal based on the changed luminance while improving contrast by independent histogram equalization according to the distribution of quantized luminance per subimage.

The basic operation of histogram equalization is to transform a given input image based on the histogram of the input image, where the histogram represents the gray level distribution in a given input image.

This histogram of gray levels provides an overall depiction of the appearance of the image. The appropriately adjusted gray level according to the sample distribution of the image improves the appearance or the contrast of the image.

Among many methods for improving the contrast, histogram equalization, which is a method for improving the contrast of a given image according to a sample distribution of an image, is most widely known and is disclosed in the following documents [1] and [2]: [1] JSLim , Two-Dimensional Signal and Image Processing, Prentice Hall, Englewood Cliffs, New Jersey, 1990, [2] RC Gonzalez and P. Wints, Digital Image Processing, Addison-Wesley, Reading, Massachusetts,

In addition, useful applications of histogram equalization methods including medical image processing and radar image processing are disclosed in [3] and [4] below: [3] J. Zimmerman, S. Pizer, E. Staab, E. Perry, W. McCartney, and B. Brenton, "Evaluation of the Effectiveness of Adaptive Histogram Equalization for Contrast Enhancement," IEEE Trans.on Medical Imaging, pp.304-312, Dec. 1988, .Wang, and DYYu, "Application of adaptive histogram equalization to x-ray chest image," Proc. of the SPIE, pp. 513-514, vol.2321, 1994.

Therefore, the technique using the histogram of a given image has been applied to various fields such as medical image processing, infrared image processing, and radar image processing.

This characteristic of the widely known histogram equalization is a drawback in practical cases. That is, since the output density of the histogram equalization is constant, the average brightness of the output image is close to the middle gray level. Actually, in histogram equalization, the average brightness of the output image is exactly the middle gray level regardless of the average brightness of the input image. Obviously, this characteristic is not desirable in practical applications. For example, a scene taken at night may appear to be too bright after histogram equalization.

In addition, the conventional histogram equalization circuit is required to have a configuration capable of storing all the occurrences of all the gray levels, thus causing a problem of high hardware cost. For example, assuming that the gray level (L) is L = 256, 256 memory elements are required to store the number of occurrences of all the gray levels, and 256 accumulators .

Therefore, according to the present invention, an input luminance image is quantized and the quantized image is divided into a predetermined number of quantized sub-images based on the average, and independent luminance histogram equalization is performed for each quantized sub-image, We propose a quantized mean split histogram equalization that can be implemented simply.

On the other hand, in order to improve the contrast, there arises a problem that the pure color signal is distorted unless the color compensation is performed on the color signal in accordance with the luminance change which occurs when a predetermined luminance process such as the above-mentioned histogram equalization is performed on the luminance signal.

For example, it is assumed that the color system is composed of Y, R-Y, and B-Y signals, and Y is changed to Y '= Y + Y by a predetermined luminance processing. The changed color signals (Y ', R-Y, B-Y) are changed to (R, G, B) values without color compensation and the resulting color signals are given by Equation (1)

[Equation 1]

R '= (R-Y) + Y'

= R +? Y

&Quot; (2) "

G '= (G-Y) + Y'

= G +? Y

&Quot; (3) "

B '= (B-Y) + Y'

= B +? Y

When Y changes to Y ', if there is no color compensation, the pure signal (R, 0,0) is mapped to (R +? Y,? Y,? Y), for example. Thus, the resulting color signal is no longer a pure signal. Similarly, if there is no color compensation, all other pure color signals are distorted.

On the other hand, Y is extracted from the input color signals R, G, and B, a correction signal is extracted from the extracted Y, and a correction signal is applied to each input R, G, and B signals as shown in Equation (1) Is disclosed in U.S. Patent No. 5,345,277. The above-mentioned patent also causes a problem that the pure color signal is distorted.

An object of the present invention is to provide a method and an apparatus for quantizing an input luminance image, dividing an input image into a predetermined number of sub-images based on the average, and independently improving histograms according to the distribution of quantized luminance per sub- And a method for improving the quality of a color signal by changing a color signal based on a changed luminance.

Another object of the present invention is to quantize an input luminance image, divide the input image into a predetermined number of sub-images based on the average, and independently improve the luminance according to the distribution of quantized luminance per sub-image to improve contrast And a circuit for improving the quality of the color signal by changing the color signal based on the changed luminance.

According to an aspect of the present invention, there is provided a method for improving image quality by processing color signals based on a luminance signal, the method comprising: (a) extracting a luminance signal from the color signals; (b) dividing the extracted luminance image signal into a predetermined number of sub-images based on the average, and independently outputting a luminance signal by histogram equalization according to the distribution of quantized luminance per sub-image; And (c) outputting the compensated color signals by varying the color signals to have the same ratio as the changed luminance signal.

According to another aspect of the present invention, there is provided an image quality improvement circuit for enhancing image quality by processing color signals based on a luminance signal, comprising: extraction means for extracting a luminance signal from the color signals; Quantized average separated histogram equalizing means for dividing the extracted luminance image signal into a predetermined number of sub-images based on the average and outputting a luminance signal that is independently histogram equalized according to the distribution of quantized luminance per sub-image to output a changed luminance signal; And color compensating means for varying the color signals to have the same ratio as the changed luminance signal and outputting compensated color signals.

1 is a diagram showing an example of quantizing an L level discrete signal into a Q level discrete signal in order to explain the concept of quantization.

2 is a diagram for explaining an interpolation concept for interpolating a quantized cumulative density function.

3 is a diagram for explaining color compensation according to variation in luminance.

4 is a diagram showing a color compensation line for preventing color saturation.

5 is a block diagram of an image quality improvement circuit according to an embodiment of the present invention.

6 is a detailed block diagram of the quantized mean separation histogram equalizer shown in FIG.

7 is a detailed circuit diagram of the color compensator shown in FIG.

8 is another detailed circuit diagram of the color compensator shown in Fig.

9 is a block diagram according to another embodiment of the image quality improvement circuit according to the present invention.

A description will be made of a method for improving the quality of a color signal using quantized mean separation histogram equalization and color compensation proposed in the present invention.

First, the quantized mean split histogram equalization algorithm will be described.

A given luminance image {X} consists of L discrete gray levels {X0, X1, ..., XL-1}, where X0 = 0 represents the black level and XL- Level. Also, Xm ∈ {X0, X1, ..., XL-1}.

Quantizing the original discrete input levels {X0, X1, ..., XL-1} to a Q discrete level defined by {Z0, Z1, ..., ZQ-1}, where ZQ- 1, and Q 1 <L, and {Z0, Z1, ..., ZQ-1} 1C {X0, X1, ..., XL-1}.

An example of quantizing the discrete signal of the L level in the Q level discrete signal is shown in Fig.

Then, Q [Xk] is called a quantization operation and is defined as follows.

Q [Xk] = Zq, if Zq-1 &lt; XK1 &lt; Zq

When {Z} = Q [{X}] and Zm = Q [Xm], Xm represents the average level of the original luminance image, {Z} represents the quantized image, and Zm represents the quantized average level, The quantized image {Z} is divided into two sub-images {Z} L and {Z} U centered at Zm. Here, all samples in the quantized sub-picture {Z} L are below the quantized average level (Zm) and all samples in the quantized sub-picture {Z} U are larger than the quantized average level (Zm).

The quantized probability density function (PDF) of each of the sub-images {Z} L and {Z} U can be expressed by the following equations (4) and (5).

&Quot; (4) &quot;

&Quot; (5) &quot;

Here, PL (Zq) is the probability of the q-th quantized gray level (Zq) in the quantized sub-picture {Z} L and PU (Zq) is the probability of the q- And NqL and NqU represent the number of times the level Zq appears in the quantized sub video Z and the quantized sub video Z. NL and NU represent the number of quantized sub video Z, L, and the total number of samples of each quantized sub-picture {Z} U.

Then, the cumulative density function (CDF) of each of the quantized sub-picture {Z} L and the quantized sub-picture {Z} U is defined by the following equations (6) and (7).

&Quot; (6) &quot;

&Quot; (7) &quot;

Here, CL (Zm) = 1 and CU (ZQ-1) = 1.

The interpolated cumulative density function cL (Xk), cU (Xk) can be roughly calculated from the quantized cumulative density functions CL (Zq), CU (Zq) through the linear interpolation shown in FIG.

Assuming that Q [Xk] = Zq 1 < Zm, Z-1 = 0 and cL (Xk) is linearly interpolated as shown in the following equation (8).

&Quot; (8) &quot;

)

Similarly, assuming Q [Xk] = Zq > Zm, cU (Xk) is linearly interpolated as shown in the following equation (9).

&Quot; (9) &quot;

Here, CU (Zm) = 0.

Based on the interpolated cumulative density function, the output (Y ') of the proposed quantized mean separation histogram equalization is given by Equation (10) for the input luminance signal Xk:

&Quot; (10) &quot;

Here, Zm '= Zm + XL-1 / (L-1), which is the next gray level after Zm in {X0, X1, ..., XL-1}.

The color compensation based on the luminance changed by the quantized mean separation histogram equalization will be described with reference to FIGS. 3 and 4. FIG.

It is assumed that C = (R, G, B) and Y, and Y is changed to Y 'by a predetermined quantized mean separation histogram equalization.

The R, G and B signals can be converted into various color signals such as (Y, I, Q), (Y, U, V) .

The relationship between the R, G, and B signals and the new color signals (Y, U, V) can be expressed by Equation (11).

&Quot; (11) &quot;

Y a11 a12 a13 R

U = a21 a22 a23 G

V a31 a32 a33 B

Here, aij is a coefficient.

On the other hand, the basic concept of color compensation according to the present invention is to change a given color from (R, G, B) space to its color direction.

First, Y can be represented by Y = a11R + a12G + a13B from the above equation (11), and (R, G, B) with a certain Y value forms a plane in the (R, G, B) space. That is, all color signals lying on a plane of Y = a11R + a12G + a13B have the same luminance value. The change in luminance from Y to Y 'implies that a given color C is shifted to the Y' plane as shown in FIG. At this time, in the present invention, it is assumed that C and C 'have the same color direction, which means that the line OC coincides with the line OC'. Thus, the compensated color C 'on the Y' plane is obtained by finding the intersection of the line OC and the Y 'plane. In summary, the original color signal C shown in FIG. 3 is mapped to C 'when the Y plane lies in the Y' plane, which is the intersection of the line OC and the Y 'plane.

Now, to find C ', (l, m, n) is defined as the directional cosine of a given color C, which is given by:

&Quot; (12) &quot;

l = R / r, m = G / r, n = B / r

here, to be.

Similarly, the direction cosine (l .m'.n ') of the output color signal C' = (R ', G', B ') can be expressed as:

&Quot; (13) &quot;

l '= R' / r ', m' = G '/ r', n '= B' / r '

here, to be. In order for these two colors to have the same color direction, the following relationship should be established as in Equation (14) or (15).

&Quot; (14) &quot;

l = l ', m = m', n = n '

&Quot; (15) &quot;

R '/ r' = R / r, G '/ r' = G / r, B '/ r'

therefore,

&Quot; (16) &quot;

&Quot; (17) &quot;

&Quot; (18) &quot;

If the relation given in the above Equations (16), (17) and (18) is substituted into Y '= a11R' + a12G '+ a13B'

&Quot; (19) &quot;

&Quot; (20) &quot;

&Quot; (21) &quot;

to be.

Therefore, the following equation (22) is obtained.

&Quot; (22) &quot;

This means that the ratio of change of luminance changes to the ratio of change of color, so that color compensation is performed by changing the color value according to the change of luminance.

Equation (16) - (18) can be expressed by the following Equation (23) - (27) using the result of Equation (22).

&Quot; (23) &quot;

[Equation 24]

&Quot; (25) &quot;

In conclusion, C 'can be obtained as follows.

&Quot; (26) &quot;

&Quot; (27) &quot;

Where k = Y '/ Y, which is the ratio between the original luminance signal and the resulting luminance signal.

Using the result of Equation (27), color compensation of other color systems can be easily performed. That is, for example, a given (Y, U, V) signal should be converted to (kY, kU, kV) as shown in the following equation (28) as a result of the color compensation given in equation do.

&Quot; (28) &quot;

&Quot; (29) &quot;

&Quot; (30) &quot;

And,

&Quot; (31) &quot;

(32)

&Quot; (33) &quot;

Now, in order to prevent color saturation due to the above-mentioned color compensation, the following color compensation is further considered.

The luminance ratio is given by k. When the color compensation is performed by the compensation lines (R ', G', B ') = k (R, G, B) shown in FIG. 4, And the maximum (Max) level is compensated with the maximum value Max, which implies saturation. This is because the color signals between the b level and the Max level are mapped to Max as a result of compensation by the compensation method described above and are not distinguished from each other.

In order to prevent this color saturation, the compensation lines (R ', G', B ') = k (R, G, B) (R ', G', B ') = A (R, G, B) + K for the color signal between level b and level Max. Here, 0 1 < 1 < 1, and A and K can be expressed by Equations (34) and (35).

&Quot; (34) &quot;

And,

&Quot; (35) &quot;

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of a color signal enhancement circuit using quantized mean separation histogram equalization and color compensation according to the present invention will be described with reference to the accompanying drawings.

5 is a block diagram of an image quality improvement circuit according to an embodiment of the present invention.

5, the luminance extractor 100 extracts the luminance signal Y from input R, G, and B signals. Here, the luminance extractor may be constituted by a matrix circuit.

The quantized average separation histogram equalizer 200 quantizes the Y signal extracted from the luminance extractor 100, divides the quantized signal into a predetermined number of quantized sub-images based on the average, And outputs a luminance signal Y 'that has been changed.

The color compensator 300 multiplies the ratio k between the Y signal output from the luminance extractor 100 and the Y 'signal output from the quantized average separation histogram equalizer 200 by the R, G, and B color signals And outputs the compensated color signals R ', G' and B '.

FIG. 6 is a detailed block diagram of the quantized mean separation histogram equalizer 200 shown in FIG.

In FIG. 6, the first quantizer 202 quantizes the input luminance image (Y = Xk's) of L discrete levels to a Q discrete level and outputs a quantized image (Zq's). The frame histogram calculator 204 calculates a gray level distribution diagram of the quantized image (Zq's) in units of one screen. Here, the screen unit may be a field, but it is a frame.

The frame average calculator 206 calculates an average level Xm of the luminance image (Y = Xk's) input in units of frames. The second quantizer 208 quantizes the average level Xm of the luminance image and outputs the quantized average level Zm.

The divider 210 divides the quantized gray level distribution diagram calculated by the frame histogram calculator 204 into a predetermined number (here, two) of quantization levels based on the quantized average level Zm output from the second quantizer 208 P (Zq) 's and PU (Zq)' s of the quantized sub-images ({Z} L, {Z} U) s), and the quantized probability density functions PL (Zq) 's and PU (Zq)' s can be calculated by Equation (4) and Equation (5).

Here, all samples in the quantized sub-picture {Z} L are below the quantized average level (Zm) and all samples in the quantized sub-picture {Z} U are larger than the quantized average level (Zm).

The first CDF calculator 212 calculates the quantized sub-picture {Z (Zq) 's) based on the probability density function PL (Zq)' s of the quantized sub-picture {Z} L, which is equal to or lower than the quantized average level Zm from the divider 210 Z} &lt; / RTI &gt; is calculated using Equation (6) above.

The second CDF calculator 214 quantizes the quantized sub-image {Z} U based on the probability density function PU (Zq) 's of the quantized sub-image {Z} U larger than the quantized average level Zm output from the divider 210 The cumulative density function (CU (Zq) 's) of the sub-image {Z} U is calculated using the above equation (7).

The CDF memory 216 stores the cumulative density functions CL (Zq) 's, CU (Zq)' s of the quantized sub-images ({Z} L, {Z} U) calculated by the first and second CDF calculators 212, and the quantized cumulative density function values CL (Zq) and CU (Zq) stored during the update are updated in the first and second interpolators 218 and 222 . Here, the synchronizing signal SYNC becomes a field synchronizing signal if the picture unit is a field, the frame synchronizing signal becomes a frame synchronizing signal, and the CDF memory 216 is used as a buffer.

The first interpolator 218 interpolates the cumulative density function value cL (Xk (z)) obtained by linear interpolation according to Equation (8) based on the cumulative density function value CL (Zq) of the quantized sub- ). Here, k = 0, 1, ..., m.

The first mapper 220 inputs the interpolated cumulative density function value cL (Xk), the input luminance signal Xk and the quantized average level Zm to obtain a sub-picture Z, which is equal to or lower than the quantized average level Zm. And maps the samples of L to the gray level from X0 to Zm according to the interpolated cumulative density function value cL (Xk).

The second interpolator 222 interpolates the cumulative density function value cU (Xk (Zq)) obtained by linear interpolation according to Equation (9) based on the cumulative density function value CU (Zq) of the quantized sub- ). Here, k = m + 1, m + 2, ..., L-1.

The second mapper 224 inputs the interpolated cumulative density function value cU (Xk), the input luminance signal Xk and the quantized average level Zm to obtain a sub-image Z } U to the gray level from Zm 'to XL-1 according to the interpolated cumulative density function value (cU (Xk)). Here, Zm '= Zm + XL-1 / (L-1).

That is, the luminance signal Xk input to the first and second interpolators 218 and 222 is compared with the cumulative density function values CL (Zq) and CU (Zq) output from the CDF memory 216, to be. However, in the present invention, the hardware is reduced by omitting the frame memory using the property of having high correlation between adjacent frames.

The comparator 226 compares the input luminance signal Xk with the quantized average level Zm output from the second quantizer 208 to generate a selection control signal. The selector 228 selects the first mapper 220 if the input luminance signal Xk is equal to or lower than the quantized average level Zm according to the selection control signal and selects the second mapper 224 otherwise, 10, i.e., a luminance signal Y 'that has been changed.

It is also possible to calculate the gray level distribution of the quantized image signals divided by the CDF calculators 212 and 214 without using the frame histogram calculator 204 without using the frame histogram calculator 204 and the CDF calculators 212 and 214 separately, To calculate the CDF.

FIG. 7 is a detailed circuit diagram of the color compensator 300 shown in FIG.

7, the arithmetic operation unit 302 calculates the ratio (k) between the Y signal output from the luminance extractor 100 of FIG. 3 and the Y 'signal output from the quantized average separation histogram equalizer 200, that is, Y' Y is calculated. Here, the computing unit can be a divider.

The first to third multipliers 304-308 multiply the input R, G and B signals by the ratio k output from the operator 302 and output the compensated R ', G' and B 'signals, respectively .

FIG. 8 is another detailed circuit diagram of the color compensator 300 shown in FIG.

8, the operator 312 calculates the ratio (k) between the Y signal output from the luminance extractor 100 of FIG. 5 and the Y 'signal output from the quantized average separation histogram equalizer 200, that is, Y' Y is calculated.

The first to third adjusters 314 to 318 compensate the R, G, and B signals input to the approximated compensation lines shown in FIG. 4 to prevent color saturation.

The first regulator 314 receives the input R signal, the ratio k output from the operator 312, ) To input the R signal as shown in Fig. 4 at a minimum level of a (= (R ', G', B ') = k (R, G, B) if the level of the input R signal is between level b, (R ', G', B ') = A (R, G, B) + K.

At this time, (R ', G', B ') = A (R', G ', B') by the value of the compensating line R, G, B) + K. In other words, = 1, the input R, G and B signals are compensated by the compensation lines (R ', G', B ') = k (R, G, B) = 0, the input R, G and B signals are compensated by the approximated compensation lines (R ', G', B ') = A (R, G, B) + K, <1, if the input R, G, B signals are at a minimum level a (= (R, G ', B') = k (R, G, B) between the a level and the maximum The approximated compensation lines (R ', G', B ') = A (R, G, B) + K compensate.

The second adjuster 314 receives the input G signal, the ratio k output from the operator 312, the parameter ) To input a G signal as shown in FIG. 4 at a minimum level of a (= (R ', G', B ') = k (R, G, B) if the level of the input G signal is between level b, (R ', G', B ') = A (R, G, B) + K.

The third arbitrator 318 receives the input B signal and the ratio (k) output from the calculator 312, the parameter ) To input the B signal as shown in FIG. 4 at a minimum level a (= (R ', G', B ') = k (R, G, B) if it is between level b and B (R ', G', B ') = A (R, G, B) + K.

9 is a block diagram according to another embodiment of the image quality improvement circuit according to the present invention.

In FIG. 9, the first color converter 400 receives the baseband digital color signals respectively denoted by R, G, and B and converts the digital luminance signals denoted by Y, U, and V into system-defined color signals.

The R, G and B signals can be converted into various color signals such as (Y, I, Q), (Y, U, V) . In the present invention, for example, (Y, U, V) is defined as a color system, Y represents a luminance signal, and U and V represent system-defined color signals, respectively. The relationship between the R, G, and B signals and the new color signals (Y, U, V) can be expressed as shown in Equation (11).

The quantized average separation histogram equalizer 500 quantizes the luminance signal Y output from the first color converter 400 and divides the quantized signal into a predetermined number of quantized sub images based on the average, And outputs the luminance signal Y 'that has been changed by histogram equalization. Here, the detailed configuration of the quantized average separation histogram equalizer 500 is the same as that shown in FIG.

The color compensator 600 receives the luminance signal Y output from the first color converter 400 and the luminance signal Y 'output from the quantized average separation histogram equalizer 500 and outputs the luminance signal Y' V) output from the first and second input / output units 400 and 400 to have the same ratio as that of the luminance signal change. Here, the color compensator 600 can output the compensated system-defined color signals U ', V' using the configuration shown in FIGS.

The second color converter 700 receives the Y 'signal output from the quantized average separation histogram equalizer 500 and the U' and V 'signals output from the color compensator 600 and outputs it to the first color converter 400 And outputs the resulting color signals R ', G', B 'through inverse conversion of the conversion performed.

That is, the resulting color signals R ', G', B 'are converted to the (R, G, B) color system from the (Y', U ', V') signal by the relationship shown in the following equation Lt; / RTI &gt;

&Quot; (36) &quot;

R 'a11 a12 a13 -1 Y'

G '= a21 a22 a23i'

B 'a31 a32 a33 V'

As described above, the present invention can be applied to a wide variety of fields related to image quality improvement of a video signal. That is, it can be applied to broadcasting equipment, radar signal processing system, medical engineering, household appliances, and the like.

The image quality improvement method and circuit of the present invention quantizes the luminance image, divides the input image signal into a predetermined number of sub-images based on the average, histologically equalizes the luminance signal independently according to the quantized luminance distribution for each sub-image, By changing the color signal on the basis of the luminance, there is an effect that sudden brightness change and artifact generated in the conventional histogram equalization can be effectively reduced to provide a color signal without distortion while improving the contrast.

In addition, the circuit of the present invention stores and accumulates only the number of occurrences of the quantized gray level, thereby simplifying the hardware and reducing the cost.

Claims (54)

A method for improving color quality by processing color signals based on a luminance signal, comprising: (a) extracting a luminance signal from the color signals; (b) quantizing the extracted luminance signal, dividing the input image into a predetermined number of sub-images based on the average, and independently histogram-equalizing the luminance image based on the quantized luminance distribution for each sub-image to output a changed luminance signal; And and (c) outputting the compensated color signals by varying the color signals to have the same ratio as the changed luminance signal. 2. The method of claim 1, wherein the compensated color signals change in the same direction as the color signals. The method of claim 1, wherein the changed luminance signal forms a luminance plane changed in the color signals space, and the luminance values of all color signals in the changed luminance plane are the same. 4. The method of claim 3, wherein the compensated color signals are obtained by an intersection of a compensated line connecting the direction of the color signals and the changed luminance plane. 5. The method of claim 4, wherein when the ratio of the extracted luminance signal and the changed luminance signal is given as k, the compensating line is given as follows. 5. The method of claim 4, wherein when the ratio of the extracted luminance signal and the changed luminance signal is k, the compensation line is given as follows. (here, ego, ego, ego, Is the maximum value) 2. The method of claim 1, wherein step (b) (b1) quantizing the level of the extracted luminance signal; (b2) obtaining an average of the extracted luminance signals on a screen basis, quantizing the obtained average, and outputting a quantized average level; (b3) dividing a signal quantized in the step (b1) into a predetermined number of quantized sub-images according to the quantized average level; (b4) obtaining an accumulated density function value based on the gray level distribution for each of the quantized sub-images; (b5) outputting an interpolated cumulative density function value for each sub-image quantized by interpolation according to the cumulative density function value obtained for each quantized sub-image; And and (b6) independently outputting a luminance signal by histogram equalization for each sub-image based on the accumulated density function value interpolated for each sub-image. 8. The method of claim 7, wherein the input image signal is divided into two sub-images according to the quantized average level in step (b3). 8. The method of claim 7, wherein the interpolation in step (b5) is linear interpolation. 8. The method of claim 7, wherein step (b6) (b61) mapping the samples of each sub-image to a gray level according to the accumulated density function value interpolated for each sub-image; (b62) comparing the extracted luminance signal with the quantized average level to generate a selection control signal; And (b63) selecting one of the signals mapped to the gray level for each sub video according to the selection control signal. The method of claim 10, further comprising: (b61 ') outputting a signal delayed by delaying the extracted luminance signal on a screen basis to step (b61). A method for improving color quality by processing color signals based on a luminance signal, comprising: (a) converting input color signals (R, G, B) into luminance signals and system-defined color signals; (b) quantizing the luminance signal, dividing the input image into a predetermined number of sub-images based on the average, independently histogram-equalizing the luminance signal based on the distribution of the luminance signals quantized for each sub-image, and outputting the changed luminance signal ; (c) varying the system-defined color signals to have the same ratio as the changed luminance signal and outputting compensated system-defined color signals; And (d) converting the modified luminance signal and the compensated system-defined color signals into compensated color signals (R ', G', B '). 13. The method of claim 12, wherein the R, G, and B signals are converted into Y, U, and V signals in step (a). 13. The method of claim 12, wherein the R, G, and B signals are converted into Y, R-Y, and B-Y signals in step (a). 14. The method of claim 12, wherein the R, G, and B signals are converted into Y, I, and Q signals in step (a). 13. The method of claim 12, wherein the compensated color signals change in the same direction as the input R, G, and B color signals. 13. The method of claim 12, wherein the changed luminance signal forms a luminance plane changed in the R, G, and B color signal spaces, and the luminance values of all color signals in the changed luminance plane are the same. 18. The image quality improvement method according to claim 17, wherein the compensated color signals are obtained by an intersection of a compensated line connecting the directions of R, G, and B color signals and the changed luminance plane. 19. The method of claim 18, wherein when the ratio of the luminance signal and the changed luminance signal is given by k, the compensation line is given by: 19. The method of claim 18, wherein when the ratio of the luminance signal and the changed luminance signal is given by k, the compensation line is given by: (here, ego, ego, ego, Is the maximum value) 13. The method of claim 12, wherein step (b) (b1) quantizing the level of the luminance signal; (b2) obtaining an average of the luminance signals on a screen basis, quantizing the obtained average, and outputting a quantized average level; (b3) dividing a signal quantized in the step (b1) into a predetermined number of quantized sub-images according to the quantized average level; (b4) obtaining an accumulated density function value based on the quantized gray level distribution map for each quantized sub-image; (b5) outputting an interpolated cumulative density function value for each sub-image quantized by interpolation according to the cumulative density function value obtained for each quantized sub-image; And and (b6) independently outputting a histogram equalized luminance signal for each of the sub-images divided by the average based on the cumulative density function value interpolated for each of the quantized sub-images. The method of claim 21, wherein in step (b3), the quantized signal is divided into two sub-images according to the quantized average level. 22. The method of claim 21, wherein the interpolation in step (b5) is linear interpolation. 22. The method of claim 21, wherein step (b6) (b61) mapping samples of each quantized sub-image to a gray level according to an accumulated density function value interpolated for each of the quantized sub-images; (b62) comparing the luminance signal with the quantized average level to generate a selection control signal; And (b63) selecting one of the signals mapped to the gray level for each of the quantized sub-images according to the selection control signal. The method of claim 24, further comprising: (b61 ') outputting a delayed signal delayed by the unit of the luminance signal to the step (b61). A circuit for processing color signals based on a luminance signal to improve image quality, the circuit comprising: Extracting means for extracting a luminance signal from the color signals; Quantizing the extracted luminance signal to divide the input image into a predetermined number of sub-images based on the quantized average, independently histogram equalizing the quantized luminance distribution for each sub-image, and outputting the changed luminance signal. Equalization means; And And color compensating means for varying the color signals to have the same ratio as the changed luminance signal and outputting compensated color signals. 27. The apparatus of claim 26, wherein the color compensation means An arithmetic unit for calculating a ratio between the extracted luminance signal and the changed luminance signal; And And a plurality of multipliers for outputting the compensated color signals by multiplying each color signal and the ratio output from the operator. 27. The apparatus of claim 26, wherein the color compensation means An arithmetic unit for calculating a ratio between the extracted luminance signal and the changed luminance signal; And The ratio output from the operator and a predetermined parameter ( And a plurality of regulators for outputting the compensated color signals by varying the value of each color signal by a predetermined compensation line. The picture quality improvement circuit according to claim 28, wherein when the ratio of the extracted luminance signal and the changed luminance signal is given by k, the compensation line is given by: The picture quality improvement circuit according to claim 28, wherein when the ratio of the extracted luminance signal and the changed luminance signal is given by k, the compensation line is given by: (here, ego, ego, ego, Is the maximum value) 27. The apparatus of claim 26, wherein the quantized mean split histogram equalization means comprises: First quantization means for quantizing the level of the extracted luminance signal and outputting a quantized signal; First calculation means for calculating an average level by inputting the extracted luminance signal on a screen basis; Second quantization means for quantizing the average level and outputting a quantized average level; Second calculation means for calculating an accumulated density function for each quantized sub-image based on a gray-level distribution diagram of a predetermined number of quantized sub-images divided according to the quantized average level; Interpolating means for interpolating the cumulative density function values of the sub-images quantized by the interpolation based on the cumulative density function values calculated for each of the quantized sub-images; And And an output unit for mapping the extracted luminance signal to a gray level according to an interpolated cumulative density function value calculated for each of the quantized sub images, and outputting a changed luminance signal. 32. The apparatus of claim 31, wherein the second calculation means A frame histogram calculator for calculating a gray level distribution of the quantized luminance signal; A divider for dividing the calculated gray level distribution into a predetermined number of quantized sub-images according to the quantized average level; And And a predetermined number of cumulative density function calculators for obtaining a cumulative density function for each quantized sub-image based on a gray-level distribution diagram of the quantized sub-image. The picture quality improving circuit according to claim 31, wherein the picture unit is a frame unit, and the predetermined number is 2. The image processing apparatus according to claim 31, further comprising: a screen memory for outputting a delayed signal by delaying the extracted luminance signal on a screen basis so as to input a signal of the same frame as the cumulative density function value calculated by the second calculation means to the interpolation means Further comprising an image quality improvement circuit. 32. The apparatus of claim 31, further comprising a buffer for updating the cumulative density function value calculated for each sub-image quantized by the second calculation means on a screen basis and for supplying the cumulative density function value stored in advance to the interpolation means during updating Wherein the image quality improvement circuit is characterized by comprising: 32. The circuit of claim 31, wherein the interpolation is linear interpolation. 32. The apparatus of claim 31, wherein the output means A first mapper for mapping the extracted luminance signal to a gray level of a first range according to an interpolated cumulative density function value corresponding thereto when the extracted luminance signal is a first sub-image of a quantized average level or lower; A second mapper for mapping the extracted luminance signal to a gray level of a second range according to an interpolated cumulative density function value corresponding thereto if the extracted luminance signal is a second sub-image that is larger than the quantized average level; A comparator for comparing the extracted luminance signal with the quantized average level to generate a selection control signal; And And a selector for selecting a first mapper if the first sub-picture is selected according to the selection control signal, and selecting a second mapper otherwise. 35. The apparatus of claim 34, wherein the output means A first mapper for mapping the delayed signal to a gray level of a first range according to an interpolated cumulative density function value corresponding thereto when the delayed signal is a first sub-image below a quantized average level; A second mapper for mapping the delayed signal to a second range of gray levels according to an interpolated cumulative density function value corresponding to the second sub-image if the delayed signal is greater than the quantized average level; A comparator for comparing the delayed signal with the quantized average level to generate a selection control signal; And And a selector for selecting a first mapper if the delayed signal is a first sub-picture according to the selection control signal, and selecting a second mapper otherwise. A circuit for processing color signals based on a luminance signal to improve image quality, the circuit comprising: First conversion means for converting input color signals (R, G, B) into luminance signals and system-defined color signals; Quantizing the luminance signal to divide the quantized image into a predetermined number of quantized sub-images based on the average, and independently quantizing histograms according to the distribution of quantized luminance per sub-image, and outputting a changed luminance signal. Discrete histogram equalization means; Color compensation means for varying the system-defined color signals to have the same ratio as the changed luminance signal to output compensated system-defined color signals; And And second conversion means for converting the changed luminance signal and the compensated system-defined color signals into compensated color signals (R ', G', B '). 40. The image quality improvement circuit according to claim 39, wherein the first conversion means converts input R, G, and B signals into Y, U, and V signals. 40. The image quality improvement circuit according to claim 39, wherein the first conversion means converts the input R, G, and B signals into Y, R-Y, and B-Y signals. The image quality improvement circuit according to claim 39, wherein the first conversion means converts the input R, G, and B signals into Y, I, and Q signals. 40. The apparatus of claim 39, wherein the color compensation means An arithmetic unit for calculating a ratio between the luminance signal and the changed luminance signal; And And a plurality of multipliers for outputting system-defined color signals compensated by multiplying each system-defined color signal and a ratio output from the operator. 40. The apparatus of claim 39, wherein the color compensation means An arithmetic unit for calculating a ratio between the luminance signal and the changed luminance signal; And The ratio output from the operator and a predetermined parameter ( ) For outputting compensated system-defined color signals by varying the value of each system-defined color signal by a predetermined compensation line. 45. The picture quality improvement circuit according to claim 44, wherein when the ratio of the luminance signal and the changed luminance signal is given as k, the compensation line is given by: 45. The picture quality improvement circuit as claimed in claim 44, wherein when the ratio of the luminance signal and the changed luminance signal is given by k, the compensation line is given by: (here, ego, ego, ego, Is the maximum value) 39. The apparatus of claim 38, wherein the quantized mean separable histogram equalization means comprises: First quantization means for quantizing the level of the luminance signal and outputting a quantized signal; First calculation means for calculating the average level by inputting the luminance signal on a screen basis; Second quantization means for quantizing the average level and outputting a quantized average level; Second calculation means for calculating an accumulated density function for each quantized sub-image based on a gray-level distribution diagram of a predetermined number of quantized sub-images divided according to the quantized average level; Interpolating means for interpolating the cumulative density function values of the sub-images quantized by the interpolation based on the cumulative density function values calculated for each of the quantized sub-images; And And an output unit for mapping the luminance signal to a gray level according to an interpolated cumulative density function value calculated for each of the quantized sub images to output a changed luminance signal. 49. The apparatus of claim 47, wherein the second calculation means A frame histogram calculator for calculating a gray level distribution of the quantized luminance signal; A divider for dividing the calculated gray level distribution into a predetermined number of quantized sub-images according to the quantized average level; And And a predetermined number of cumulative density function calculators for obtaining a cumulative density function for each quantized sub-image based on a gray-level distribution diagram of the quantized sub-image. The picture quality improving circuit according to claim 47, wherein the picture unit is a frame unit and the predetermined number is 2. The image processing apparatus according to claim 47, further comprising a screen memory for delaying the luminance signal by a screen unit and outputting a delayed signal for inputting a signal of the same frame as the cumulative density function value calculated by the second calculation unit to the interpolation unit Wherein the image quality improvement circuit is characterized by comprising: The image processing apparatus of claim 47, further comprising a buffer for updating the cumulative density function value calculated for each sub-image quantized by the second calculation unit on a screen basis and for supplying the cumulative density function value stored in advance to the interpolation means during updating Wherein the image quality improvement circuit is characterized by comprising: 48. The circuit of claim 47, wherein the interpolation is linear interpolation. 49. The apparatus of claim 47, wherein the output means A first mapper for mapping the luminance signal to a gray level of a first range according to an interpolated cumulative density function value corresponding to the luminance signal if the luminance signal is a first sub- A second mapper for mapping the luminance signal to a second range of gray levels according to an interpolated cumulative density function value corresponding thereto if the luminance signal is a second sub-image that is larger than the quantized average level; A comparator for comparing the luminance signal with the quantized average level to generate a selection control signal; And And a selector for selecting a first mapper if the first sub-picture is selected according to the selection control signal, and selecting a second mapper otherwise. 51. The apparatus of claim 50, wherein the output means A first mapper for mapping the delayed signal to a gray level of a first range according to an interpolated cumulative density function value corresponding thereto when the delayed signal is a first sub-image below a quantized average level; A second mapper for mapping the delayed signal to a second range of gray levels according to an interpolated cumulative density function value corresponding to the second sub-image if the delayed signal is greater than the quantized average level; A comparator for comparing the delayed signal with the quantized average level to generate a selection control signal; And And a selector for selecting a first mapper if the delayed signal is a first sub-picture according to the selection control signal, and selecting a second mapper otherwise.