JPH03126377A - Gradation correction device - Google Patents

Abstract

PURPOSE:To attain gradation correction even to a signal of a bright portion or an intermediate brightness level by providing a histogram arithmetic circuit, a histogram limiter, an adder, a subtractor, a clip circuit, an accumulation control register circuit and a normalized control register circuit. CONSTITUTION:A mean value, a mode value, a maximum value, a minimum value, a deviation coefficient, a black area and a white area are detected from a brightness distribution data in a histogram memory 8 by a histogram arithmetic circuit 15. Then a histogram limiter.adder.subtraction.clip circuit 16 generates a corrected histogram data and an accumulation control register circuit 17 varies a start and stop brightness level in response to a video signal to control the range of accumulation. Moreover, a normalized control register circuit 18 controls a maximum level of a luminance signal outputted from an accumulation histogram memory 10. Thus, sufficient gradation is corrected even to a signal of a bright portion or an intermediate brightness level.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a gradation correction device that can be used to correct the gradation of video signals of television receivers, video tape recorders, and the like.

Conventional technology In recent years, as color television receivers have become larger,
In order to make the image appear clearer, the gradation of the video signal is corrected by passing the video signal through a non-linear amplifier, and C
In order to expand the dynamic range of images on R'I', emphasis has been placed on gradation correction devices.

Hereinafter, a conventional gradation correction device as described above will be described with reference to the drawings. FIG. 5 shows a block diagram of a conventional tone correction device. In FIG. 5, reference numeral 1 denotes a black detection circuit which detects a dot portion of the input luminance signal and outputs a black detection signal. A gain control circuit 2 controls the gain of the black detection signal using a gain control voltage and outputs an amplified black detection signal. 3 is an addition circuit that adds the input luminance signal and the amplified black detection signal and outputs one output luminance signal. 4 is a black peak output circuit that holds the black peak level of the output luminance signal and outputs the voltage as a black peak hold voltage to a comparator circuit 5; Compare voltages.

The operation of the tone correction device configured in this way will be described below. FIG. 6 shows signals from various parts of a conventional tone correction device.

First, the input luminance signal a is input to the black detection circuit 1, and the black signal in the portion of the luminance signal below a certain value is extracted and output as a black detection signal. Next, the black detection signal S is input to the gain control circuit 2, and the gain B') is controlled according to the gain control voltage f, and the amplified black detection signal C
The output is as follows.

This signal C is then input to the adder circuit 3, where it is added to the input luminance signal a, and an output luminance signal d whose dynamic range on the output side has been expanded is output. This signal d is simultaneously input to the black peak hold circuit 4. This black peak hold circuit 4 detects the blackest luminance signal level and outputs the voltage as a black peak hold voltage e. Comparison circuit 5 compares this black peak hold voltage e with reference m pressure g generated from voltage source 6, and feeds back the difference to gain control circuit 2 as gain control voltage f. This feedback system becomes stable when the black peak hold voltage e becomes equal to the reference voltage g. In this way, when there is a detected black component, the black peak hold voltage e is always controlled to be equal to the reference voltage g.
The dynamic range on the output side is expanded and gradation correction is performed.

FIG. 7 shows a block diagram of another conventional tone correction device.

In FIG. 7, 7 is an AD converter that converts the input luminance signal into a digital value. 8 is a histogram memory that takes the input luminance (the luminance distribution of the number), and in the previous calculation, the luminance is taken at the address of the memory, and the frequency is stored in that data. 9 is the cumulative value of the data in the histogram memory 8. 10 is a cumulative histogram memory that stores the cumulative result of the histogram cumulative adder 9. Generally, the brightness is stored in the address of the memory, and the frequency is stored in the data. A lookup table calculation circuit normalizes each data based on the cumulative histogram data so that its maximum cumulative frequency becomes the maximum value of the output luminance signal.12 is the normalized data by the lookup table calculation circuit 11. This is a lookup table memory that stores data, and generally the input brightness level is taken at the address of the memory, and the output brightness level is stored in that data. 14 is a DA converter that converts into a value.
This is a timing control circuit that controls the timing of No. 13.

The operation of the tone correction device configured in this way will be described below. FIG. 8 is a graph showing the state of luminance conversion in this conventional gradation correction device.

First, the input luminance signal a is input to the AD converter 7, converted into a digital value, and output as a converted input luminance signal h. This converted input luminance signal is stored in the histogram memory 8.
For each piece of input data, 1 is added to the data at that address. The contents of this memory are cleared at regular intervals and each data is set to O. This period is generally selected to be one vertical scanning period or an integral multiple thereof. This situation is shown in FIG. 8(a).

Next, the histogram cumulative adder 9 sequentially reads out the histogram memory data i from the histogram memory 8, cumulatively adds them, and stores the resulting cumulative histogram data j in the cumulative histogram memory 10. This situation is shown in FIG. 8(b).

Next, the lookup table calculation circuit ■ calculates a normalization coefficient such that the maximum cumulative value of the cumulative histogram memory data stored in cumulative histogram memory January 0 becomes the maximum value of the output brightness level, and calculates a normalization coefficient that is multiplied by each data in the cumulative histogram memory 10, and the resulting lookup table data l is stored in the lookup table memory 12. This situation is shown in FIG. 8(C).

The lookup table memory 12 reads the data of the converted input luminance signal as an address, and the data at that address becomes the converted output luminance signal m obtained by the above calculation, and this converted output luminance signal m is L) By converting it into an analog signal by the A converter 13, an output luminance signal d is obtained.

This situation is shown in FIG. 8(d).

Problems to be Solved by the Invention However, in the above configuration, in the example shown in FIG. There was a problem in that correction could not be made and the dynamic range could not be sufficiently improved.

The example shown in Figure 7 is generally called histogram flattening processing, and the dynamic range is expanded to 100%, but if this processing is performed on a normal TV show video, This has the problem that the image is far from the real image and is rather difficult to see.

The present invention solves the above-mentioned problems, and makes it possible to perform sufficient gradation correction not only on the output side but also on bright areas and intermediate brightness level signals, prevent the dynamic range from expanding too much, and improve fidelity. An object of the present invention is to provide a gradation correction device that can perform high-contrast gradation correction.

Means for Solving the Problems In order to solve the above problems, the gradation correction device of the present invention includes a histogram memory, a histogram calculation circuit, a histogram limiter/addition/subtraction/clip circuit, an accumulation control register circuit, and a normalization control register circuit. ,
Histogram cumulative addition circuit, cumulative histogram memory,
The histogram calculation circuit calculates the average value, mode value, maximum value, minimum value, and deviation (dispersion) coefficient of the input luminance signal from the data in the histogram memory. , black area, and white area, and calculates each control data of limiter, clear addition, subtraction, accumulation start, accumulation stop, and maximum brightness level from these, and the histogram limiter, addition, subtraction, and clip circuit calculates the control data based on the control data. Creates corrected histogram data by replacing the values above or below a certain value with a certain value (limiter, clip), adding or subtracting a certain value to the data in the histogram memory, and performs cumulative control. The register circuit controls the range in which the corrected histogram data is accumulated by setting a start luminance level at which the accumulation starts and a stop luminance level at which the accumulation is stopped in the histogram accumulation/addition circuit. The system is configured to control the maximum value data of the cumulative histogram memory for normalizing the data in the memory, and perform gradation correction.

Operation With the above configuration, when the histogram memory detects the luminance distribution of the input video signal and performs the commonly performed histogram flattening process described in the conventional example, the feature of high dynamic range of this histogram flattening process is In order to optimize this effect depending on the video, the average value of the input luminance signal,
Mode value, maximum value, minimum value, deviation coefficient, black area, white area 5
The histogram limiter, addition, subtraction, and clipping circuits use these data to replace the data in the histogram memory with values above or below a certain value (limiter, clip), or is added or subtracted to create corrected histogram data to optimize the effect, and when taking the cumulative histogram of the corrected histogram data in the histogram accumulation adder circuit, the start brightness level at which the accumulation starts from the accumulation control register circuit number. The stop luminance level at which accumulation is stopped is changed according to the video signal to control the range of accumulation, and also when normalizing the cumulative histogram memory data output from the cumulative histogram memory using a lookup table calculation circuit. By controlling the maximum level of the output luminance signal output from the cumulative histogram memory 10 using the normalization control register circuit, automatic control of brightness and contrast, known as conventional ABL and ACL, can be performed at the same time. In this way, the effect of histogram flattening processing can be optimized by adding or subtracting a certain value from the histogram, or replacing values above or below a certain value with a certain value. You can control the range when taking a histogram,
Tone correction is performed by controlling the maximum value during normalization.

EXAMPLE Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 shows a block diagram of a tone correction device according to an embodiment of the present invention. In FIG. 1, 7 is an AD converter, 8 is a histogram memory, 9 is a histogram cumulative adder, 10 is a cumulative histogram memory, 11 is a lookup table calculation circuit, 12 is a lookup table memory, and 13 is a D
A converter 14 is a timing control circuit, which is the same as in the conventional example.

15 is a histogram calculation circuit, which includes an average brightness level calculation circuit that calculates the average value of the input brightness signal from the data in the histogram memory 8, a mode brightness level detection circuit that detects the mode value, and a maximum value that detects the maximum value and the minimum value. Detection circuit, minimum value detection circuit, histogram memory 8
It has an arithmetic circuit that calculates a deviation (dispersion) coefficient from the data, an arithmetic circuit that calculates a white area and a black area, etc., and based on the results obtained from these, the histogram limiter/addition/subtraction/clipping circuit 16, which will be described later, Each control data of the cumulative control register circuit 17 and the normalization control register circuit 18 is determined, and each circuit, 16-
Output to 1B. In addition, smoothing operations (smoothing using an average filter or median filter) are performed on the histogram data by taking the average or median of the local area of the conversion input luminance signal data and replacing it with the central value. Eliminate signals and reduce conversion errors due to noise.

Here, the average value of the luminance signal is calculated from the data in the histogram memory 8 according to equation (1).

Average value = input brightness level x histogram value ÷ total number of pixels
.....Furthermore, the mode value is the input luminance signal level having the maximum value of the histogram value. The maximum value is the upper limit input brightness level of the histogram distribution, and the minimum value is the lower limit input brightness level of the histogram distribution.

Further, the black area refers to the number of pixels in which the input luminance level of the histogram distribution is in the range of, for example, 0 to 40%, and the white area refers to the number of pixels in which the input luminance level of the histogram distribution is in the range of, for example, 60% or more. The deviation coefficient may be obtained by calculating the standard deviation or simply using equation (2).

Deviation coefficient = (maximum value - minimum value)
From the deviation coefficient, white area, black area, limiter, clip,
Calculates control data for addition, subtraction, cumulative start, cumulative stop, and maximum brightness level.

Histogram limiter/addition/subtraction/clip circuit 16
is a histogram memory 8 and a histogram cumulative addition circuit 9
A limiter circuit that fixes the frequency above a certain level to a constant value, a clip circuit that fixes the frequency below a certain level to zero, and a limiter circuit that fixes the frequency above a certain level to zero for the data in the histogram memory 8. , an adder and a subtractor that add and subtract a constant level constant or a non-constant level constant to histogram data, and a subtractor that subtracts a constant number of pixels from the dark and low brightness sides of the histogram distribution. The histogram cumulative addition circuit is configured such that part or all of each level or the number of pixels of the circuit is varied by control data of limit values, clip values, addition values, and subtraction values given from the histogram calculation circuit 15. Correct the histogram memory data input to 9.

Note that the added value here may be a constant value or a non-constant value. In addition, subtraction values include subtracting a constant value or non-constant value from all histogram data, subtracting the number of pixels equal to the number of subtraction values from the lower limit of the histogram distribution, and subtracting the number of pixels equal to the number of subtraction values from the upper limit of the histogram distribution. Subtraction has three types of subtraction values or some of them.

The accumulation control register circuit 17 has registers for setting the luminance level at which the cumulative addition is started and the luminance level at which the cumulative addition is stopped when performing cumulative addition in the cumulative histogram, and has registers for setting the luminance level at which the cumulative addition starts and the luminance level at which the cumulative addition is stopped. Each level of the register is varied by the accumulation stop, and a histogram accumulation adder circuit 9 that performs accumulation calculation is controlled according to the value.

When the normalization control register circuit 18 normalizes the data in the cumulative histogram memory 10 and creates a look-up table, the normalization control register circuit 18 receives the maximum brightness level of the normalized output brightness signal from the histogram calculation circuit 15;
The normalization coefficient in the lookup table calculation circuit 11 is controlled according to the value.

The operation of the gradation correction device configured as described above will be explained below. Figure 2 shows the operation of each part.

First, the input luminance signal a is input to the AD converter 7, converted into a digital value, and the converted signal is output as a converted input luminance signal. The histogram memory 8 uses this converted input luminance signal as a memory address and adds 1 to the data at that address. By performing this operation for one vertical period, the histogram distribution of the input luminance signal can be obtained. This is shown in FIG. 2(a).

Next, the histogram calculation circuit 15 reads out the data p in the histogram memory 8 containing this histogram distribution, performs a smoothing calculation in the manner described above, removes the singular signal, re-records it in the histogram memory 8, and inputs it. Calculate the average value, mode value, maximum value, minimum value, deviation coefficient, white area, and black area of the luminance signal.Then, from each of these calculated values, the fixed vine level, clip level, addition value, subtraction value, and cumulative calculation are calculated. A start brightness level, a stop brightness level, and a maximum brightness level after normalization are determined, and these control data 0 are applied to a limiter/addition/subtraction/clip circuit 16, an accumulation control register circuit 17,
Transfer to normalization control register 18.

Next, the limiter/addition/subtraction/clip circuit 16 outputs the smoothed data i from the histogram memory 8.
is read out, and the previously transferred limiter level, clip level, addition value, and subtraction value are calculated for each data, and the corrected histogram data n obtained as a result is calculated.
is output to the histogram cumulative addition circuit 9 (see Fig. 2(b)
)). Here, for example, if the added value is an added value as shown in FIG. 3(C), the cumulative addition result will have the conventional r correction characteristic as shown in FIG. 3(d). In addition, in the case of constant value addition, the larger the added data, the more
The data becomes more linear, and the smaller it is, the closer it is to a histogram flattening process.

Then, the histogram cumulative addition circuit 9 generates a cumulative histogram of the corrected histogram data n in the range from the cumulative start brightness level to the cumulative stop brightness level using the cumulative start brightness level and cumulative stop brightness level signals ψ given from the cumulative control register circuit 17. Data j is calculated and the result is stored in cumulative histogram memory 10.

Next, the lookup table calculation circuit 11 reads the cumulative histogram data k obtained in the above manner from the cumulative histogram memory 10, and outputs the maximum output of the lookup table whose maximum cumulative data is given by the normalization control register circuit 18. A normalization coefficient is determined so that the brightness level becomes r, and this normalization coefficient is calculated for all cumulative histogram data, and the result is stored in the lookup table memory 12. At this time, the address is the input luminance level, and lookup table data l, which is the calculation result, is stored in that data. In other words, when the average brightness level is high, the maximum output brightness level is lowered, and when the average brightness level is low, the maximum output brightness level is increased.
It can operate like automatic brightness control (ABL). This operation is shown in FIG. 2(e).

Next, the look-up table memory 12 in which the look-up table data l has been set reads the data of the converted input luminance signal as the address of this memory, and reads the data as in the conventional example. As shown in FIG. 2(f), the conversion output q [degree signal m]
l) Output to A converter 13. And IJA converter 1
3 converts the converted output luminance signal +'1n into the analog output luminance signal d and outputs it.

The timing control circuit 14 controls the operation of each circuit so that each section operates at the timings described above.

As described above, according to this embodiment, the histogram flattening process is applied to the gradation correction of a video device such as a television set, and the correction amount is calculated based on the average value, mode value, and input luminance signal obtained from the histogram data. Maximum value, minimum value, black area, white area,
By controlling using deviation coefficients, etc., it is possible to perform sufficient gradation correction not only on the output side as in the past, but also on bright areas and intermediate brightness level signals, and prevent the dynamic range from expanding too much. , it is possible to perform more faithful and high-contrast gradation correction.

FIG. 2 is a diagram explaining the operation when the added value is a constant value, while FIG. 3 is a diagram explaining the operation when the added value is set to a non-constant value.

FIG. 4 shows a block diagram of a tone correction device according to another embodiment of the present invention. In FIG. 4, 7 is an AD converter, 8 is a histogram memory, 12 is a look-up table memory,
13 is an IJA converter, and 14 is a timing control circuit, which are the same as in the conventional example.

Reference numeral 19 denotes an arithmetic unit composed of a microcomputer, and a histogram memory is a part of the memory of the microcomputer, and the microcomputer reads the histogram data and performs various arithmetic operations as described in the previous embodiment. perform an operation that integrates the That is, first, the histogram data p is smoothed using an average value filter, a median filter, etc., and the histogram data i resulting from this calculation is applied to the input luminance signal's average value, mode value, maximum value, minimum value, deviation coefficient, etc. Detects white area, black area, etc., determines a limiter value, addition value, subtraction value, cumulative start brightness level, cumulative stop brightness level, and output maximum brightness level from these data, performs cumulative calculation of histogram data i, From this, lookup table data l is determined and the result is stored in lookup table memory 12.

The operation of the tone correction device configured in this way will be described below. The input luminance signal a is AD converted, and its histogram distribution is stored in the histogram memory 8. The process up to this point is the same as the conventional example. The arithmetic unit 19 calls this data i, performs the above-mentioned arithmetic operations, obtains lookup table data k, and stores it in the lookup table memory 12. The subsequent operation is similar to the conventional example. Note that the timing of the entire operation is controlled by the timing control circuit 14.

As described above, according to this embodiment, the arithmetic unit (microcomputer) performs arithmetic operations for each part at once, making it possible to perform the same gradation correction as in the previous example with fewer circuit configurations. can.

Effects of the Invention As described above, according to the present invention, the gradation correction device includes a histogram memory, a histogram calculation circuit, a histogram limiter/addition/subtraction/clip circuit, a product stiffness control register circuit, a normalization control register circuit, and a histogram. By providing a cumulative addition circuit, a cumulative histogram memory, a lookup table calculation circuit, a lookup table memory, and a timing control circuit, it is possible to control not only the output side of the gradation of the video signal but also bright areas and intermediate purity levels. It is possible to provide a gradation correction device that can sufficiently perform gradation correction on signals, prevent the dynamic range from expanding too much, and perform more faithful and high-contrast gradation correction.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a gradation correction device according to an embodiment of the present invention, Fig. 2 is a characteristic diagram for explaining the operation of each part in Fig. 1, and Fig. 3 shows a non-constant addition value in Fig. 1. 4 is a block diagram of a gradation correction device according to another embodiment of the present invention. FIG. 5 is a block diagram of a conventional gradation correction device. Figure 6 is a waveform diagram of each part of Figure 5, and Figure 7 is a waveform diagram of each part of Figure 5.
This figure is a block diagram of another conventional tone correction device, and FIG. 8 is a characteristic diagram for explaining the operation of each part of FIG. 7. 8... Histogram memory, 9... Histogram cumulative addition circuit, 10... Cumulative histogram memory, 11
... Lookup table calculation circuit, 12... Lookup table memory, 14... Timing control N
15... Histogram calculation circuit, 16... Limiter/addition/subtraction/clip circuit, 17... Accumulation control register circuit, 18... Normalization control register circuit, 19... Arithmetic unit, i ... Histogram memory data, k... Cumulative histogram memory data, 1... Lookup table data, n... Correction histogram data, 0... Control data,
r...Maximum output brightness data. 21st

Claims (1)

[Claims] 1. A histogram memory that stores the histogram of the video luminance signal of a television receiver, video tape recorder, video camera, video disk, etc.; a histogram accumulator that cumulatively adds the data in this histogram memory; and a histogram cumulative adder that stores the cumulative addition results. a look-up table calculation circuit that normalizes the data of the cumulative histogram memory by the maximum value data of the cumulative histogram memory; and a look-up table memory that stores the results; The average value, mode value, maximum value, minimum value, deviation (dispersion) coefficient, black area, and white area of the input luminance signal are detected from the data, and from this, the limiter, clip, addition, subtraction, etc.
A histogram calculation circuit calculates each control data of cumulative start, cumulative stop, and maximum brightness level, and control data from this histogram calculation circuit is used to control the data in the histogram memory by a certain value or more or less than a certain value. A histogram limiter/addition/subtraction/clip circuit that adds or subtracts a certain value to create corrected histogram data, and control data from the histogram calculation circuit determines the start luminance level at which the histogram accumulation and addition circuit starts accumulation. and an accumulation control register circuit that sets a stop luminance level at which accumulation is stopped and controls a range in which the correction histogram data is accumulated, and control data from the histogram calculation circuit to normalize the data in the accumulation histogram memory. A gradation correction device equipped with a normalization control register circuit that controls the maximum value data of the cumulative histogram memory.