JPH09224174A - Adaptive gamma correction device using integral lookup table - Google Patents

Abstract

(57) [Summary] (Correction) [Problem] To determine gamma correction from a complicated distribution histogram in real time without increasing the scale of hardware. A distribution histogram is used to integrate a luminance signal. After comparing the distribution level of this brightness signal with the matrix unit 60 that receives the R, G, B image signals and outputs the brightness signal, the number of distributions corresponding to each level is counted to obtain a distribution histogram of the brightness signal. A histogram extraction unit 62, a histogram integration unit 64 that accumulates and integrates the luminance signal distribution histogram, an integration lookup table that corrects the input video signal using the integration curve, and an output correction signal. And a gamma determining unit 68 for gamma-processing and outputting the final luminance signal.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a CC for a video camera.
The present invention relates to a gamma correction device for correcting a luminance signal output from a D (charge coupled device), and more specifically, a look-up table (hereinafter referred to as a LUT) that integrates a distribution histogram of the luminance signal output from a CCD of a video camera. ) Is used to efficiently use the dynamic range of the signal.

[0002]

2. Description of the Related Art Gamma means the inclination of a straight line portion when input / output characteristics in a photoelectric conversion system or an electro-optical conversion system such as a video camera are displayed on the X and Y axes. On a TV, the logarithmic ratio of the luminance of the subject to the luminance of the cathode ray tube according to the output voltage of the camera is displayed as gamma.

The gamma correction device is composed of a non-linear circuit in which the output signal satisfies an exponential curve with respect to the level of the input signal. FIG. 1 is a block diagram showing a conventional gamma correction apparatus, showing a Neuro gamma correction apparatus. The conventional gamma correction apparatus shown in FIG. 1 uses R, G, B of image signals.
A matrix unit 10 for receiving a signal and extracting a luminance signal;
A feature amount extraction unit 12 that extracts a feature amount of a luminance signal using the luminance signal output from the matrix unit 10, and a feature amount output from the feature amount extraction unit 12 (a dark portion, an intermediate portion,
Gamma determination unit 1 that determines gamma using bright areas)
4. The gain improving unit 16 improves the gain by using the luminance signal output from the matrix unit 10 and the gamma output from the gamma determining unit 14.

The operation of such a conventional neuro-gamma correction device is as follows. The matrix section 10 extracts a luminance signal from the input R, G, B signals. The extracted luminance signal is inspected for the level of the luminance signal, and the number of dark portions, intermediate portions and bright portions of the luminance signal included in one frame is calculated. Here, when it is determined that the backlight condition is present, gamma processing is performed using a method of compressing the luminance level of image data of low occurrence frequency included in a frame and expanding the luminance level of image data of high occurrence frequency. Therefore, a method of expanding the dynamic range of the luminance signal is used.

When a subject is photographed in a backlit state, the amount of incident light increases in the bright part of the image and the white clip region also expands, but the amount of incident light decreases in the dark part, making it impossible to express the gradation of the image. Become. That is, since it is difficult to distinguish slightly dark image data from very dark image data, they are displayed in the same darkness. In this case, 10
Even if the dynamic range of 0% is used, the amount of signal data distributed in the intermediate luminance portion is small. Actually, most of the signal data is concentratedly distributed in the low brightness part and the high brightness part.

In order to solve the above-mentioned problem, a neuro gamma correction device has been adopted in the prior art. FIG. 2 is a graph showing the distribution characteristics of the luminance signal when the subject is backlit.
With reference to FIG. 2, the range of the luminance signal is divided into three equal parts and divided into low luminance, intermediate luminance and high luminance, and then the image data is corrected in one step. That is, when the distribution density of the luminance signal is high, the input signal is expanded up to 3 times, but when the distribution density of the luminance signal is low, the input signal is compressed.

FIG. 3 is a graph showing the relationship between input / output characteristics and image correction in the backlight condition. Referring to FIG.
An image in a normal light state has an input / output characteristic index curve of a 1: 1 bypass line, but has an input / output characteristic curve like a straight line “α” under special photographing conditions such as a backlit state.
When performing the primary correction to flatten the luminance signal distribution,
An input / output characteristic curve such as "β" is obtained.

Further, when two-step image correction is performed after one-step image correction, the input / output characteristic curve shown in FIG. 4 is obtained as a result. Such correction is performed by the gain improving unit 16 shown in FIG. 1, and the conventional neuro-gamma correction device adjusts the degree of gain improvement according to the feature amount of the luminance component. However, the conventional neuro-gamma correction apparatus processes the feature amount of the image data into three equal parts according to the distribution of the luminance component, that is, divides into a dark part, an intermediate part and a bright part, and processes the result, as shown in FIG. The backlight correction function is not performed when the distributed luminance signal is distributed. This is because the luminance signal is distributed between the low luminance portion and the intermediate luminance portion, or is distributed between the high luminance portion and the intermediate luminance portion, so that there is a relative difference in distribution between the respective portions. This is because it is not clear.

Further, as shown in FIG. 5B, when the difference in the distribution of the luminance signal is 3 or more, the image correction work is not smoothly performed. Therefore, in order to solve the above-mentioned problems from the conventional device, it is necessary to divide the current three brightness levels into tens of levels of division and gain improving portions. However, it is very difficult to increase the number of signal processing steps due to real-time operation and size limitation of hardware.

[0010]

The present invention was created to solve the above-mentioned problems, and is an LUT capable of appropriately correcting image data of a subject having an uneven histogram of a composite luminance signal. It is an object of the present invention to provide an adaptive gamma correction device using the.

[0011]

In order to achieve the above object, a gamma correction apparatus according to the present invention includes a matrix unit for receiving a R, G, B image signal and outputting a luminance signal, and a matrix unit for outputting the luminance signal. After comparing the distribution level of the luminance signal, a histogram extraction unit for obtaining the distribution histogram of the luminance signal by counting the number of distributions corresponding to each level,
A histogram integrator that accumulates and integrates the luminance signal distribution histogram output from the histogram extractor,
An integration LUT that corrects the input video signal using the integration curve obtained from the histogram component unit, and a gamma determination unit that gamma-processes the correction signal output from the integration LUT to output the final luminance signal. It is characterized by including and.

[0012]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. FIG. 6 is a block diagram showing the configuration of the adaptive gamma correction apparatus according to the present invention.
Referring to FIG. 6, the gamma correction apparatus uses R, R
A matrix unit 60 that receives the G and B signals and outputs a luminance signal, a histogram extraction unit 62 that extracts a distribution histogram of the luminance signal using the luminance signal output from the matrix unit 60, and a histogram output from the histogram extraction unit 62 Histogram integrator 6 that integrates using
4. An integration LU that corrects the image signal input using the output of the histogram integration unit 64 into an image suitable for the current screen
And a gamma determining unit 68 that determines a gamma from the integration LUT 66 and outputs a corrected image.

FIG. 7 is a detailed view of the histogram extraction unit 62 shown in FIG. 6, which is composed of 256 comparators 70 and 256 counters 72 corresponding to the comparators 70, respectively. When the comparator 70 compares the levels according to the level of the luminance signal output from the matrix unit 60 shown in FIG. 6 and if they match, the corresponding value of the counter 72 is increased. That is, the level of the video signal in one frame is compared with each reference value to extract the number of pixels corresponding to each level.

FIG. 8 is a detailed diagram of the histogram integrator 64 shown in FIG. 6, in which the number of pixels output from the counter 72 shown in FIG. 7 is changed from the low level counter (0). The signals are sequentially output to the high level counter (255) and accumulated and integrated by the accumulator 80 shown in FIG. On this occasion,
The value accumulated according to the output of each counter is stored in the integration LUT 66 shown in FIG.

FIG. 9 is a graph showing the luminance distribution of the image signal in the backlit state according to the present invention, and FIG. 10 is the integral LU of FIG.
7 is a graph showing input / output characteristics of T. Next, the operation of the present invention will be described with reference to FIGS. In the case of the histogram extraction unit 62 which extracts the distribution histogram of the luminance signal using the comparator, the bright part and the dark part are extremely contrasted on the screen of one frame under the special image condition such as the backlight condition. Since there is almost no image data in the middle part, the characteristics of the extracted histogram are shown in FIG.
As shown.

The integrating LUT 66 shown in FIG. 10 provides the corrected input / output characteristics of one input image and the DRA
A device for obtaining input / output characteristics is configured using M, SRAM, FRAM, EEPROM and the like. That is, the gamma correction apparatus according to the present invention counts the levels of the luminance signal separately by the conventional neuro-gamma correction apparatus, and then, in the case of the backlight condition, a certain amount of the input luminance signal is determined according to the level of the input luminance signal. Unlike the method of correcting the image by improving the gain, the LUT obtained from the distribution of the luminance signal
Can be used to adaptively obtain more appropriate input / output characteristics.

Referring to the input / output characteristics shown in FIG. 10, a dark light signal of 60 or less as the input signal Y has a high distribution of luminance signals in the distribution of the entire luminance signal between 0 and 255. Therefore, the range is enlarged so that the details are displayed. Further, referring to the histogram shown in FIG. 9, since there is no luminance signal at the luminance signal level between 60 and 200, even if the input signal changes drastically, the output does not change.

Referring to the input / output characteristic curve shown in FIG. 10, when a luminance signal between 200 and 255 is input,
The output of the luminance signal is formed at a ratio that is relatively amplified according to the influence of the luminance signal. As a result, the amount of change in the luminance signal is displayed finely in the portion where the distribution of the input luminance signal is high, and in the portion where the distribution is low or where there is no distribution, it is possible to use the dynamic range efficiently. .

Further, since the gamma correction apparatus according to the present invention uses the method by integration, it is possible to use the method shown in FIGS. 5 (A) and 5 (B).
An appropriate operation can be performed even with the complicated and non-uniform luminance signal distribution shown in FIG. The output flattened in accordance with the image correction operation as described above is corrected using a gamma curve as shown in FIG.

[0020]

As described above, according to the present invention, the integration method for making the dynamic range efficient is different from the compression / expansion of the input / output characteristics of the conventional neuro-gamma correction device in that the compression / expansion range is The input / output characteristic curve is appropriately obtained even with the compositely distributed luminance signal level, which makes backlight compensation impossible with the neuro-gamma corrector because it is greatly increased.

In addition, it is difficult for the conventional gamma correction apparatus to obtain an input / output characteristic curve from the luminance signal levels distributed in a complex manner due to the enormous increase in the scale of hardware and the difficulty of real-time processing. According to the invention,
It is possible to obtain the input / output characteristic curve without causing problems such as increase in the scale of additional hardware and real-time processing.

[Brief description of drawings]

FIG. 1 is a block diagram showing a conventional gamma correction device.

FIG. 2 is a graph showing a distribution characteristic of a luminance signal at the time of backlight photography.

FIG. 3 is a graph showing a relationship of input / output characteristics with respect to image correction in a backlit state.

FIG. 4 is a graph showing input / output characteristics after ideal image correction.

5A and 5B are graphs showing an example in which backlight compensation cannot be performed by a conventional gamma correction device.

FIG. 6 is a block diagram showing a configuration of an adaptive gamma correction device according to the present invention.

FIG. 7 is a detailed diagram of a histogram extraction unit shown in FIG.

FIG. 8 is a detailed view of the histogram integration unit shown in FIG.

FIG. 9 is a graph showing a luminance distribution of an image signal in a backlit state according to the present invention.

10 is a graph showing input / output characteristics of the integration LUT of FIG.

[Explanation of symbols]

 60 matrix part 62 histogram extraction part 64 histogram integration part 66 integration LUT 68 gamma determination part 72 counter

Claims (5)

[Claims] 1. A gamma correction apparatus, which compares a distribution level of a luminance signal output from the matrix unit with a matrix unit that receives the R, G, and B image signals and outputs the luminance signal, and then corresponds to each level A histogram extraction unit that obtains a distribution histogram of luminance signals by counting the number of distributions, a histogram integration unit that accumulates and integrates the luminance signal distribution histogram output from the histogram extraction unit, and obtains from the histogram integration unit An integration look-up table that corrects the input video signal using the obtained integration curve; and a gamma determining unit that gamma-processes the correction signal output from the integration look-up table and outputs a final luminance signal. An adaptive gamma correction device comprising: 2. The histogram extracting unit includes a plurality of comparators and a counter corresponding to the comparators, compares the levels of video signals in one frame, and extracts the number of pixels corresponding to each level. The adaptive gamma correction device according to claim 1, wherein 3. The adaptation according to claim 2, wherein the histogram integrator sequentially accumulates and integrates values of the low level counter and the high level counter. Type gamma correction device. 4. The integration lookup table is RAM
The adaptive gamma correction apparatus according to claim 1, wherein the input / output characteristics are corrected by using. 5. The integration look-up table is EEP.
The adaptive gamma correction apparatus using the integral look-up table according to claim 1, wherein the input / output characteristics are corrected using a ROM.