JPH10271409A - Image quality improvement circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the image quality improvement circuit that applies simple processing to a signal with a high quality characteristic. SOLUTION: An average luminance level (APL) measurement section 1 calculates a signal APL denoting an average luminance signal level of a luminance signal Y for one image pattern. An arithmetic section 2 conducts nonlinear signal processing that suppresses a signal level of the luminance signal Y whose level is less than the APL by using a correction curve generated by an accumulation arithmetic operation of a specific value g1 (g1>1.0) and that expands a signal level of the luminance signal Y whose level is more than the APL by using a correction curve generated by an accumulation arithmetic operation of a specific value g2 (g1<1.0). Then the luminance signal Yn whose gradation is corrected is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image quality improving circuit suitable for realizing gradation correction while maintaining a natural feeling.

[0002]

2. Description of the Related Art With the development of digital signal processing technology, advanced signal processing, which is difficult to achieve with conventional analog signal processing, has become possible. Various digital signal processing technologies have also been devised for improving the quality of television images. I have.

Among these digital signal processing techniques for improving image quality, a technique for performing nonlinear signal processing on a luminance signal of an image as a technique for improving sharpness and gradation correction of an image while maintaining a natural feeling is proposed. The following prior arts are known.

(1) The average luminance level of a luminance signal is measured,
When this is high, the white level is extended, and when it is low, gamma correction is applied.

(2) A method in which one screen is divided into several areas, and a correction curve is selected for each area from a correction curve table set in advance based on the average luminance level of each area.

(3) A method of measuring the distribution form of the signal level of the luminance signal and expanding the gradation in a portion where the frequency is concentrated.

[0007] Among these prior arts, the method (1) has a problem that the signal processing is simple, but the correction function does not work for an image whose average luminance level is an intermediate level. In addition, in the method (2), there is a possibility that correction is performed using a different correction curve in each area, and there is a concern that the continuity of the luminance signal at the boundary portion is lost and image quality is deteriorated.

On the other hand, the method (3) has a great effect of tone correction, but is useful for signal processing of luminance distribution measurement and extraction of this characteristic parameter (detection of a region where frequencies are concentrated and spread of distribution). There is a problem that a huge amount of calculation is required.

For this reason, it is desired to realize an image quality improvement circuit which is simple in signal processing and has high quality while maintaining a natural feeling.

[0010]

SUMMARY OF THE INVENTION An object of the present invention is to solve the problems of the prior art, to achieve simple signal processing, and
An object of the present invention is to provide a high quality image quality improvement circuit which maintains a natural feeling.

[0011]

In order to achieve the above object, the present invention employs the following technical means of signal processing. That is, first, an average luminance signal level (hereinafter abbreviated as APL) of a luminance signal of one screen image (one frame period or one field period of an image signal) is calculated.

Next, non-linear signal processing is performed on the luminance signal of one screen with a correction curve generated by a power operation using the calculated APL as a representative point. That is, the specific value g1 (g1> 1.0) for a luminance signal whose signal level is lower than APL.
Performs nonlinear signal processing to suppress the signal level with the correction curve generated by the power operation of. For a luminance signal whose signal level is equal to or higher than APL, nonlinear signal processing for extending the signal level by a correction curve generated by a power operation of a specific value g2 (g2 <1.0) is performed.

FIG. 7 schematically shows the non-linear processing of a luminance signal in the above-mentioned prior art (3). First, the frequency of occurrence of each signal level (minimum level Ymin to maximum level Ymax) of the luminance signal is measured, and the luminance distribution characteristics shown in FIG. Next, this luminance distribution characteristic is analyzed,
Extract characteristic parameters such as a region where frequencies are concentrated and a spread of distribution. Then, a luminance conversion table of the correction curve shown in the figure is created by using the characteristic parameters, and non-linear signal processing for enlarging the gradation of a portion where the frequency is concentrated is performed. Therefore, it is possible to perform gradation correction in a form suitable for each image (the image A and the image B shown in the figure). On the other hand, there is a problem that an enormous amount of calculation is required for signal processing in order to perform accurate gradation correction.

FIG. 8 shows an outline of the nonlinear processing of a luminance signal according to the present invention. In the luminance distribution characteristics shown in FIG. 3, the APL calculated according to the present invention is located at the center of gravity of this distribution. In addition, in the majority of general images, a region where frequencies are concentrated (a region where the frequency of occurrence is high in the figure) tends to occur before and after the APL. In the present invention, focusing on these viewpoints, the APL is set as a representative point, and the correction generated by the exponentiation operation shown in FIG. The nonlinear signal processing of the luminance input Y is performed using the curve.

That is, for a luminance signal Y having a signal level lower than APL, a luminance output Yn with gradation corrected is obtained by a correction curve of a characteristic 1 by a power operation of a specific value g1 (g1> 1.0) shown in Expression 1. .

[0016]

Yn = (APL−Ymin) ｛(Y−Ymin) / (APL−Ymin)｝ ** g1 + Ymin ** indicates a power. (Equation 1) On the other hand, for a luminance signal whose signal level is equal to or higher than APL. The
As shown in Equation 2, the gradation-corrected luminance output Yn is obtained by the correction curve of the characteristic 2 by the power operation of the specific value g2 (g2 <1.0).
Get.

[0017]

Yn = (Ymax-APL) ｛(Y-APL) / (Ymax-APL)｝ ** g2 + APL (Equation 2) Note that the characteristic value g1 is larger than 1.0 and g2 is larger than 1.0. Although the relationship is small, g1 is about 1.2 to 1.4, and g2 is 0.8.
By setting the value to about 0.9, it is possible to achieve good gradation correction while maintaining a natural feeling for most images.

Also, unlike the measurement of the frequency of occurrence in the prior art (3), the calculation of APL in the present invention can be obtained by simply integrating the luminance signal levels over one screen and calculating the average value. It can be realized by extremely simple signal processing.

Further, regarding the non-linear signal processing based on the correction curves of the exponentiation operation shown in the characteristics 1 and 2, software processing by a DSP or a microprocessor, or A
A luminance conversion table created in advance using the PL as a parameter is stored in the ROM, and the luminance conversion table can be easily realized by a table lookup calculation process or the like.

As described above, according to the signal processing of the technical means of the present invention, it is possible to realize an image quality improvement circuit that performs high-quality characteristic gradation correction while maintaining a natural feeling by simple signal processing.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION

(Embodiment 1) A first embodiment of the present invention will be described with reference to FIGS. In all the embodiments, the luminance signal is assumed to be a digital signal sequence sampled at the sampling frequency fs.

FIG. 1 is this block diagram. The APL measuring section 1 calculates an average luminance signal level of the input luminance signal Y over one screen, and outputs this as a signal APL.

The arithmetic unit 2 performs a non-linear signal processing of the luminance signal with the correction curves of the characteristic 1 and the characteristic 2 based on the power operation using the signal APL as a representative point, and outputs a luminance signal Yn with gradation corrected.

Next, FIG. 2A shows an example of the configuration of the APL measuring section in the present embodiment, and FIG.
Shown in The gate section 3 opens the gate when the control signal CS1 is High, as shown in FIG. 3B, and outputs a luminance signal in the period of one line effective pixel area. The integrator 4 performs an operation of clearing the content of the integrator to zero at the beginning of one frame period in which the control signal CS2 is High. After that, the integration operation is performed during the Low period. Then, at the end of one frame period, the integrated result of the luminance signal of the effective pixel area in one frame (one field) period is output. The division unit 5 performs a division operation at the end of one frame period in which the control signal CS3 is High. That is, an average luminance signal level of one screen is calculated by an operation of dividing the result of integrating the luminance signals by the total number of effective pixels in one frame (one field) period, and this is output as a signal APL. The control unit 6 generates control signals CS1 to CS3 necessary for these operations.

FIG. 3 shows an example of the configuration of the calculation unit in this embodiment. The arithmetic control unit 12 calculates the input luminance signal Y and the signal AP
The signal level is compared with the signal level L, and control signals MOD, g, OFS necessary for performing a power operation based on the characteristic 1 when Y <APL and the characteristic 2 when Y> = APL are output.

The subtractor 7 determines that Y <APL (control signal MOD is
In the case of (Low), the subtractor 7-1 outputs the result of the subtraction operation of Y-Ymin and the subtractor 7-2 outputs the result of the subtraction operation of APL-Ymin as signals S1 and S2, respectively. When Y> = APL (the control signal MOD is High), the subtractor 7-1 outputs YA
The PL and subtraction unit 7-2 outputs the result of the subtraction operation of Ymax-APL as signals S1 and S2, respectively. The division unit 8 performs an operation of dividing S1 by S2. The exponentiation operation unit 9 calculates Y
In the case of <APL, the input signal is g1 (control signal g = g
1) A power operation for raising is performed. In the case of Y> = APL, a power operation for raising the input signal to the power of g2 (control signal g = g2) is performed. The multiplier 10 performs a multiplication operation on the input signal and the signal S2. The adder 11 adds a control signal OFS (OFS = Ymin) to the input signal when Y <APL. Y> = A
In the case of PL, the control signal OFS (OFS = A
PL). Then, a luminance signal Yn obtained by performing gradation correction on this output with a correction curve based on exponentiation is obtained.

It should be noted that these signal processings can be realized by software processing using a DSP or a microprocessor.

As described above, according to the present embodiment,
With a simple signal processing, it is possible to realize an image quality improvement circuit with high quality characteristics that performs gradation correction suitable for the luminance distribution of an image while maintaining a natural feeling. Then, a remarkable effect can be obtained for improving the image quality of the television image.

(Embodiment 2) Next, a second embodiment of the present invention will be described with reference to FIGS.

FIG. 4 is a diagram showing an example of this block configuration. AP
The L measurement unit 1 calculates an average luminance signal level of the input luminance signal Y over one screen and outputs this as a signal APL.

The setting unit 13 performs an operation of assigning a plurality of representative levels according to the value of the signal APL. That is, as shown in FIG. 9, the signal APL having a value of Ymin to Y1 corresponds to the representative level Ya (Ya = (Y1 + Ymin) / 2), Y
1 to Y2 are representative levels Yb (Yb = (Y1 + Y2)
/ 2), those of YK to Ymax are representative levels Yk (Yk =
(YK + Ymax) / 2) is assigned. Then, a signal TBL (TBL1, TBL2, TBL) indicating the representative level is specified.
BLk). In order to perform accurate gradation correction, the number of representative levels needs to be increased.
If it is set to about 6, there is no practical problem.

As shown in FIG. 9, the ROM operation unit 14 includes a luminance conversion table TBL based on a correction curve generated by a power operation using representative levels Ya, Yb, and Yk as representative points.
1, TBL2, and TBLk. A corresponding table is selected by a signal TBL, a table lookup operation is performed using the selected table, and a luminance signal Yn with gradation corrected is output.

As described above, according to the present embodiment,
With an even simpler signal processing than in the first embodiment, it is possible to realize an image quality improvement circuit having high quality and performing gradation correction suitable for the luminance distribution of an image while maintaining a natural feeling. Then, a remarkable effect can be obtained for improving the image quality of the television image.

In the above embodiment, the average luminance signal level over one screen is used as the signal APL. Therefore, the number of pixels required for integration is enormous, for example, 768 pixels × 480 lines in an NTSC image signal. On the other hand, if the average luminance signal level is obtained for each line, the number of pixels required for integration can be greatly reduced to 768 pixels. Therefore, if the signal APL can be calculated from the average luminance signal level obtained for each line, signal processing can be further simplified.

FIG. 10 shows an example of measuring the luminance level of various images (images A to F). In the figure, A
PL is the average luminance signal level over one screen, MAX and M
IN indicates the maximum value and the minimum value of the average luminance signal level obtained for each line, and (MAX + MIN) / 2 indicates the average value of the maximum value and the minimum value. In the images other than the image C, the value of APL and the average value (MAX + MIN) / 2 are substantially equal. Therefore, this average value (MAX + MIN) / 2 can be used as an average luminance signal level over one screen. Further, although there is a slight difference between the two in the image C, the difference is such that it can be almost absorbed in the process of assigning the representative level in the second embodiment.

(Embodiment 3) The third embodiment of the present invention has been made from the above viewpoint and will be described below with reference to FIGS. 5 and 6. FIG.

FIG. 5 is a block diagram showing the configuration, which comprises an APL simple measurement unit 15, a setting unit 13, and a ROM operation unit 14. Note that the setting unit 13 and the ROM calculation unit 14
The description is omitted because it is the same as the embodiment.

FIG. 6 shows an example of the configuration of the APL simple measuring section in this embodiment. The gate unit 3 opens the gate when the control signal CS1 is High, as in the first embodiment, and
A luminance signal for the period of the line effective pixel area is output.

The line-by-line APL measuring section 16 clears the contents of the integrator to zero at the beginning of each line where the control signal CS4 is high, and performs an integration operation in the low period thereafter. Then, at the end of the one-line period, the integrated value is divided by the number of effective pixels of one line, and a signal of an average luminance signal level for each line is output.

The MAX detector 17 detects that the control signal CS2 is Hi.
Set the maximum value to Ymin at the beginning of one frame period of gh,
In the subsequent periods, magnitude comparison with the input average luminance signal level in line units is sequentially performed, and the maximum value in one frame period is detected. The MIN detector 18 sets the minimum value to Ymax at the beginning of one frame period in which the control signal CS2 is High, and sequentially compares the level with the input average luminance signal level in line units in the subsequent period. , 1
The minimum value in the frame period is detected.

The averaging section 19 adds and averages the output signals of the MAX detection section 17 and the MIN detection section 18 at the end of one frame period when the control signal CS3 is High, and outputs the result to the signal AP.
Output as L. The control unit 20 generates control signals CS1 to CS4 similar to those in FIG.

As described above, according to the present embodiment,
With an even simpler signal processing than in the first and second embodiments, it is possible to realize an image quality improvement circuit having high quality characteristics that performs gradation correction suitable for the luminance distribution of an image while maintaining a natural feeling. Then, a remarkable effect can be obtained for improving the image quality of the television image.

As a method for simplifying the signal processing, in addition to the above-described method, the average luminance signal level is reduced by a luminance signal sequence in which the number of sampling points is reduced to 1 / n by n: 1 sub-sample signal processing. There is a way to measure. The fourth described below
The sixth to sixth embodiments are made from this viewpoint.

(Embodiment 4) FIG. 11 is a block diagram of a fourth embodiment of the present invention, which is realized by adding a sub-sampling section 21 to the first embodiment.

The sub-sampling section 21 performs n: 1 sub-sampling signal processing of the input luminance signal Y,
The number of sample points in the vertical direction or in the horizontal and vertical directions is 1 / n
And generates a reduced luminance signal sequence. Then, the APL measurement unit 1 calculates an average luminance signal level over one screen by using the luminance signal sequence, and outputs this as a signal APL.

(Embodiment 5) FIG. 12 is a block diagram of a fifth embodiment of the present invention, which is realized by a configuration in which a sub-sampling section 21 is added to the second embodiment.

As in the fourth embodiment, the sub-sampling section 21 generates a luminance signal sequence in which the number of sample points is reduced by n: 1 sub-sampling signal processing. Then, the APL measurement unit 1 calculates an average luminance signal level over one screen by using the luminance signal sequence, and outputs this as a signal APL.

(Embodiment 6) FIG. 13 is a block diagram of a sixth embodiment of the present invention, which is realized by a configuration in which a sub-sampling section 21 is added to the third embodiment.

As in the fourth embodiment, the sub-sampling section 21 generates a luminance signal sequence in which the number of sample points is reduced by n: 1 sub-sampling signal processing. Then, the APL simple measuring unit 15 calculates the average value of the maximum value and the minimum value of the average luminance signal level of each line of the luminance signal sequence, and outputs this as the signal APL.

As described above, according to the fourth to sixth embodiments, the luminance distribution of the image can be improved while maintaining a natural feeling with a smaller circuit scale than the first to third embodiments. An image quality improvement circuit having high quality characteristics for performing appropriate gradation correction can be realized. In addition, a remarkable effect can be obtained for cost reduction and high quality television images.

It is known that, when watching a television image, the area to be watched is a roughly 1/2 area centered on the center of the screen. Therefore, as shown in FIG. 14, it is also possible to cut out a substantially half area at the center of the screen as a window area, and to use the signal APL calculated in this area to perform tone correction mainly on the gaze area. As a result, it is possible to realize an image quality improvement circuit adapted to the viewing situation and simplifying the signal processing. The seventh to ninth embodiments described below are made from this viewpoint.

(Embodiment 7) FIG. 15 is a block diagram of a seventh embodiment of the present invention, which is realized by a configuration in which a window setting section 22 is added to the first embodiment.

As shown in FIG. 14, the window setting section 22 generates a luminance signal sequence included in the window area at the center of the screen among the input luminance signals Y. Then, the APL measurement unit 1 calculates an average luminance signal level in the window area from the luminance signal sequence, and outputs this as a signal APL.

(Embodiment 8) FIG. 16 is a block diagram of an eighth embodiment of the present invention, which is realized by a configuration in which a window setting unit 22 is added to the second embodiment.

The window setting section 22 generates a luminance signal sequence included in the window area of the input luminance signal Y as in the seventh embodiment. Then, the APL measurement unit 1 calculates an average luminance signal level in the window area from the luminance signal sequence, and outputs this as a signal APL.

(Embodiment 9) FIG. 17 is a block diagram of a ninth embodiment of the present invention, which is realized by adding a window setting unit 22 to the third embodiment.

The window setting section 22 generates a luminance signal sequence included in the window area of the input luminance signal Y, as in the seventh embodiment. Then, the APL simple measuring unit 15
Calculates the average value of the maximum value and the minimum value of the average luminance signal level of each line of the window area in this luminance signal sequence,
This is output as signal APL.

As described above, according to the seventh to ninth embodiments, it is possible to realize an image quality improvement circuit adapted to the viewing situation and simplifying the signal processing. In addition, a remarkable effect can be obtained for cost reduction and high quality television images.

The quantization accuracy of the luminance signal of an image is 8 bits in most cases. Therefore, in either embodiment,
It is desirable to calculate APL with 8-bit accuracy. However, for further simplification, it is also possible to execute with an accuracy of about 6 bits.

[0060]

According to the present invention, it is possible to realize a high quality image quality improvement circuit for performing tone correction suitable for the luminance distribution of an image while maintaining a natural feeling by simple signal processing.
In addition, cost reduction and high quality television images can be achieved.

[Brief description of the drawings]

FIG. 1 is a block diagram of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of an APL measurement unit in the first embodiment.

FIG. 3 is a block diagram of a calculation unit according to the first embodiment.

FIG. 4 is a block diagram of a second embodiment of the present invention.

FIG. 5 is a block diagram of a third embodiment of the present invention.

FIG. 6 is a block diagram of an APL simple measurement unit according to a third embodiment.

FIG. 7 is a characteristic diagram of a luminance signal non-linear process according to a conventional technique.

FIG. 8 is a characteristic diagram of the luminance signal nonlinear processing according to the present invention.

FIG. 9 is a diagram showing input / output characteristics of a ROM operation unit in the second and third embodiments.

FIG. 10 is an explanatory diagram of measurement of a luminance level in various images.

FIG. 11 is a block diagram of a fourth embodiment of the present invention.

FIG. 12 is a block diagram of a fifth embodiment of the present invention.

FIG. 13 is a block diagram of a sixth embodiment of the present invention.

FIG. 14 is an explanatory diagram of a luminance average level measurement using a window.

FIG. 15 is a block diagram of a seventh embodiment of the present invention.

FIG. 16 is a block diagram of an eighth embodiment of the present invention.

FIG. 17 is a block diagram of a ninth embodiment of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... APL measurement part, 2 ... Calculation part, 3 ... Gate part, 4 ... Integration part, 5 ... Division part, 6 ... Control part.

Claims (6)

[Claims] An image quality improving circuit for improving image quality by nonlinear signal processing of a luminance signal of an image, wherein an average luminance signal level of the luminance signal for one screen of an image, that is, for one frame period or one field period of the image signal is determined. Means for calculating, and means for performing non-linear signal processing using a correction curve generated by a power operation on the luminance signal of one screen, wherein the means for non-linear signal processing uses the average luminance signal level as a representative point, A specific value g1 (g1> 1.
0), the signal level is suppressed to a value equal to or lower than the luminance signal level by a correction curve generated by the exponentiation operation, and a specific value g2 is set for the luminance signal having a level equal to or higher than the average luminance signal level.
An image quality improving circuit for performing a non-linear signal processing for extending a signal level to a value equal to or higher than the luminance signal level with a correction curve generated by a power operation of (g2 <1.0). 2. The non-linear signal processing means according to claim 1, further comprising a setting unit for setting a plurality of representative levels corresponding to the level value of the average luminance signal level, and a power having the representative level obtained by the setting unit as a representative point. The image quality improvement circuit according to claim 1, wherein nonlinear signal processing is performed using a correction curve generated by calculation. 3. A plurality of tables in which an input luminance signal and an output luminance signal obtained by non-linear processing based on a correction curve generated by exponentiation are associated with a plurality of tables for each of the representative points of the respective representative levels. And selecting one of the plurality of tables stored in the ROM according to the representative level obtained by the setting unit, and performing a non-linear signal processing of the luminance signal by a lookup table calculation process using the selected table. The image quality improvement circuit according to the item. 4. The method of calculating an average luminance signal level of a luminance signal, wherein a line average luminance signal level is measured in units of one scanning line period of the luminance signal, and one frame period or one field period of the line average luminance signal level is measured. 4. The image quality improvement circuit according to claim 1, wherein a maximum value and a minimum value are detected, and an average value of the maximum value and the minimum value is set as an average luminance signal level. 5. A luminance signal sequence in which the number of sample points is reduced to 1 / n in a horizontal direction, a vertical direction, or a horizontal / vertical direction by performing signal processing of n: 1 sub-sampling on the luminance signal. 5. The image quality improvement circuit according to claim 1, wherein an average luminance signal level is calculated using the luminance signal sequence. 6. An outline 1 centered on the center of the screen on one screen.
6. The image quality improvement circuit according to claim 1, wherein a window area having an area of / 2 is set, and an average luminance signal level is calculated based on luminance signals included in the window area.