KR101024058B1 - Image contrast control method and device for local dimming backlight - Google Patents

Abstract

)을 추출한다. 그런 다음, 분할부(108)가 상기 Y-Cb-Cr 영상의 크기를 M * N 서브 블록 크기로 분할하면, 상기 제 1 계산부(110) 및 제 2 계산부(112)는 상기 분할된 서브 블록별로 각 블록별 휘도 평균값(

) 및 표준편차 정보(σ_b)를 계산한다. 선택부(116)에 의해 상기 분할된 서브 블록 중 상기 블록별 휘도 평균값(

)이 상기 전체 휘도 평균값(

)보다 작은 서브 블록이 선택되면, 제어부(118)는 상기 블록별 휘도 평균값(

) 의 표준편차(σ_b) 내로 한정하여 상기 전체 휘도 평균값(

) 이하의 값을 가지는 픽셀에만 국부적인 히스토그램 확장 기법(local histogram stretching)을 적용하여 컨트라스트를 향상시킨다. 아울러, 상기 히스토그램 확장이 수행된 각 서브 블록의 정보를 기초로 하여 백라이트 제어부(120)는 상기 픽셀의 광원에 공급되는 전류를 제어한다. 그와 같은 본 발명에 따르면, 블록 효과의 제거로 인하여 영상의 컨트라스트가 개선되고, 일부 픽셀에 대한 히스토그램 확장 수행으로 인해 하드웨어 구성이 간단해지며 전력소모가 감소하는 이점이 있다. The present invention relates to a method and apparatus for controlling image contrast for local dimming backlight. According to the present invention, the first extractor 104 extracts a luminance Y component from a Y-Cb-Cr image converted from an RGB image, and the second extractor 106 extracts an overall luminance average value (

). Then, when the division unit 108 divides the size of the Y-Cb-Cr image into M * N sub-block sizes, the first calculation unit 110 and the second calculation unit 112 divide the size of the divided sub. The average luminance value of each block for each block

) And standard deviation information (σ _b ) are calculated. The luminance average value of each block among the divided subblocks by the selecting unit 116 (

Is the overall luminance average value (

When a sub block smaller than) is selected, the controller 118 determines the luminance average value of each block (

The overall luminance average value (limited to the standard deviation σ _b ) of

Local histogram stretching is applied only to pixels having the following values to improve contrast. In addition, the backlight controller 120 controls the current supplied to the light source of the pixel based on the information of each sub block in which the histogram expansion is performed. According to the present invention, the contrast of the image is improved due to the removal of the block effect, and the hardware configuration is simplified and the power consumption is reduced due to the histogram expansion for some pixels.

Description

Method for controlling image contrast for local dimming backlight and its device {Method for controlling a Image Contrast for Local Dimming Backlight and Apparatus

The present invention relates to image contrast control. In particular, in a display device in which a backlight LED is disposed on a rear surface of a liquid crystal and a local dimming method is applied, the input image is divided into block units of the same size, and the contrast is independently determined in the block units. A method and apparatus for controlling image contrast for local dimming backlight for controlling.

Currently, various methods for improving contrast of image information input from a display apparatus have been proposed.

Among them, a method using a histogram of an image having various applications in the image processing field will be described.

First, a histogram equalization (also referred to as a global HE) method is a representative method of improving contrast of an input image using a histogram of an image. The method is a method of rearranging the contrast values of the image based on the probability distribution of the contrast values in the input image. This method provides simple and relatively good performance for various images. However, the method improves the entire input image to the same level, and is not suitable for an input image having various brightness characteristics. In other words, if the contrast is extended by performing histogram smoothing when a pixel having a specific brightness is dominant in one image information, a loss may occur in contrast in the case of arbitrary pixels.

Thus, a local histogram equalization method has been proposed in which the input image is divided into sub-blocks and performs histogram smoothing for each region. Since the method divides and processes the input image, the dynamic region of each divided image can be separated and expanded, thereby preventing the problem of causing interference between a region having a large contrast value and a small region. However, when the input image is divided into sub-blocks and processed, a blocking effect phenomenon occurs due to a difference in histogram smoothing functions between adjacent blocks. That is, the block effect encodes each block independently, causing discontinuity in the edge portion of the block, causing deterioration in image quality. Therefore, there is a demand in the image processing field to reduce the block effect.

A block-overlapped histogram equalization method for overlapping blocks by pixel and a partial overlapped sub-block histogram smoothing method for partially overlapping as a method of removing the block effect. Block Histogram Equalization method has been proposed.

However, the above methods have a disadvantage in that the hardware configuration is complicated and the amount of computation is large. This problem causes a problem that energy is inefficiently consumed in the display device. In particular, the portable display device is important to extend the battery life time, and as a result, efforts to reduce the power consumption of the device has been continuously performed. Although it can improve the power consumption, it is not suitable for the goal of reducing power consumption. The power consumption is applied not only to portable devices but also to general display devices.

As a result, when the input apparatus divides and processes the input image into sub-blocks, a new method for minimizing power consumption while maximizing contrast by eliminating block effects is needed.

Accordingly, an object of the present invention is to solve the above problems, and to improve the contrast of image information by improving the block effect generated when the input image is divided and processed.

Another object of the present invention is to minimize the power consumption in the display device.

)을 계산하는 단계; 상기 전체 휘도 평균값(

)이 계산되면 상기 YCbCr 영상의 크기를 M * N 서브 블록 크기로 분할하는 단계; 상기 분할된 서브 블록별로 각 블록별 휘도 평균값(

) 및 표준편차 정보(σ_b)를 계산하는 단계; 상기 분할된 서브 블록 중에서 상기 블록별 휘도 평균값(

)이 상기 전체 휘도 평균값(

)보다 작은 서브 블록만을 선택하는 단계; 그리고 상기 선택된 서브 블록 중 상기 블록별 휘도 평균값(

) 의 표준 편차(σ_b) 내로 한정하여 상기 전체 휘도 평균값(

) 이하의 값을 가지는 픽셀에만 국부적인 히스토그램 확장 기법(local histogram stretching)을 적용하는 단계를 포함하여 구성된다. According to a feature of the present invention for achieving the above object, converting an RGB image to a YCbCr image; Extracting a luminance (Y) component from the Y-Cb-Cr image; Overall luminance average value of the Y-Cb-Cr image

Calculating; The overall luminance average value (

Is calculated, the size of the YCbCr image is divided into M * N subblock sizes; The luminance average value of each block for each of the divided subblocks (

) And the standard deviation information σ _b ; The luminance average value of each block among the divided subblocks (

Is the overall luminance average value (

Selecting only subblocks smaller than); And a luminance average value of each block among the selected subblocks (

The overall luminance mean value (limited to the standard deviation σ _b ) of

And applying local histogram stretching only to pixels having a value less than or equal to).

Controlling a current supplied to a light source of the pixel based on information of each sub block in which the histogram expansion is performed; And integrating the image divided into the sub-blocks and converting the image into an RGB image.

)이 클수록 상기 히스토그램 확장이 증가하도록 제어하는 것을 특징으로 한다.The local histogram expansion technique controls the degree of expansion of the histogram in consideration of the distribution of pixels located in the sub block, wherein the standard deviation of the sub block is small and the luminance average value of each block of the sub block (

Is increased so that the histogram expansion increases.

The unselected sub-blocks in the selecting step may maintain the luminance value of the original image.

The local histogram extension technique may allow only dark areas to be darker than some of the original images for some pixel values of the selected subblock.

)를 추출하는 제 2 추출부; 상기 Y-Cb-Cr 영상정보의 크기를 M * N 서브 블록 크기로 분할하는 분할부; 상기 분할된 서브 블록별로 각 블록별 휘도 평균값(

) 을 구하는 제 1 계산부 및 표준편차정보(σ_b)를 구하는 제 2 계산부; 상기 Y-Cb-Cr 영상정보의 휘도 평균값(

)과 상기 각 블록별 휘도 평균값(

)의 크기를 비교하는 비교부; 상기 비교 결과에 따라 상기 분할된 블록 중에서 히스토그램 확장기법을 적용할 서브 블록을 선택하는 선택부; 그리고 상기 선택된 서브 블록 중 다음 식(1)에 해당하는 픽셀들에 대해서만 다음 식(2)에 의해 국부적인 히스토그램 확장 기법(local histogram stretching)을 수행하는 제어부를 포함하여 구성된다. According to another aspect of the present invention for achieving the above object, a first extraction unit for extracting the luminance (Y) component of the converted Y-Cb-Cr image information from the input RGB image; The luminance average value of the Y-Cb-Cr image information (

A second extracting unit for extracting; A divider dividing the size of the Y-Cb-Cr image information into M * N sub-block sizes; The luminance average value of each block for each of the divided subblocks (

A first calculation unit for obtaining _" ) and a second calculation unit for obtaining standard deviation information [sigma] _b ; Luminance average value of the Y-Cb-Cr image information (

) And the luminance average value of each block (

Comparing unit for comparing the size of; A selection unit for selecting a sub-block to which the histogram expansion technique is applied among the divided blocks according to the comparison result; And a control unit performing local histogram stretching by only the pixels corresponding to the following equation (1) among the selected subblocks by the following equation (2).

- σ_b <

<

+ σ_b ----- 식(1)

-σ _b <

<

+ σ _b ----- Equation (1)

----- 식(2)

----- Formula (2)

는 히스토그램 확장된 출력 픽셀의 휘도 값,

는 각 서브블록의 입력 픽셀의 휘도 값, α는 확장계수,

는 각 블록별 휘도 평균값, σ_b는 표준편차정보이다. here,

Is the luminance value of the histogram extended output pixel,

Is the luminance value of the input pixel of each subblock, α is the expansion coefficient,

Is an average value of luminance for each block, and sigma _b is standard deviation information.

<

을 만족하는 서브 블록을 상기 히스토그램 확장기법을 적용할 서브 블록으로 선택한다.The selection unit

<

A sub block satisfying the P 1 is selected as a sub block to which the histogram extension technique is applied.

And a backlight controller configured to control a supply current supplied to a light source of the pixel based on the information of each sub block on which the histogram expansion is performed.

According to the present invention, the histogram is extended only for subblocks in which the average luminance value of each region is smaller than the total luminance average value in the state where the average luminance value of the input image, the luminance average value and the standard deviation of each region of the original image divided into a predetermined number are obtained. In this case, the histogram expansion technique is applied to only pixels having a value less than or equal to the overall luminance average value of the original image among pixels located in the sub-block. The block effect that has occurred is removed.

In addition, the histogram expansion technique is used for only some pixels of the divided sub-blocks, thereby simplifying image processing and processing speed of the display device, thereby simplifying the hardware configuration.

In addition, the display device adjusts the current supplied to the LED light source based on the information of each sub-block in which the histogram expansion is performed, thereby reducing power consumption and improving additional contrast.

Hereinafter, a method and an apparatus for controlling image contrast for local dimming backlight according to the present invention will be described in detail with reference to a preferred embodiment shown in the accompanying drawings.

An embodiment of the present invention is applied to a display device to which an LCD panel structure capable of realizing local dimming by placing a backlight LED behind a light guide plate, wherein the display device is a general device. Will be omitted.

1 is a block diagram of an image contrast control apparatus according to a preferred embodiment of the present invention.

Referring to FIG. 1, the image contrast control apparatus 100 includes a converter 102 that converts RGB image information into a Y-Cb-Cr image.

The first extractor 104 extracts a luminance (Y) component from the Y-Cb-Cr image converted by the converter 102. In general, the equation used to extract the luminance component is as follows.

Y = 0.257R + 0.504G + 0.098B + 16 ----- Formula (1)

Cb = -0.148R-0.291G + 0.439B + 128 ----- Formula (2)

Cr = 0.439 R-0.368 G-0.071 B + 128 ----- Formula (3)

)를 추출하는 제 2 추출부(106)가 구비된다.Average Luminance of whole image in the Y-Cb-Cr image

) Is provided with a second extraction unit 106.

A divider 108 is provided for dividing the Y-Cb-Cr image into M * N predetermined number of blocks.

)을 계산하는 제 1 계산부(110)와, 블록별 표준편차(σ_b)를 계산하는 제 2 계산부(112)가 구비된다.Average Luminance of sub-blocks of each block (called 'sub-blocks') divided by the divider 108 into M * N size (Average Luminance of sub-blocks,

The first calculation unit 110 for calculating a) and the second calculation unit 112 for calculating the standard deviation (σ _b ) for each block is provided.

)과 상기 제 1 계산부(110)에 의해 계산된 블록별 휘도 평균값(

)의 크기를 비교하는 비교부(114)가 구비된다. The luminance average value extracted by the second extraction unit 106 (

) And the luminance average value for each block calculated by the first calculator 110 (

Comparing unit 114 for comparing the size of the) is provided.

<

을 만족하는 서브 블록만을 선택한다. 이는 분할된 서브 블록 중에서 충분히 밝은 픽셀들로만 구성된 서브 블록에 대해서는 본 실시 예의 히스토그램 확장 기법을 적용하지 않고자 함이다. As a result of the comparison, a selector 116 is provided to select a subblock to which the histogram extension technique is applied from the divided subblocks. The selection unit

<

Only subblocks satisfying are selected. This is to avoid applying the histogram extension technique of the present embodiment to a subblock composed of only sufficiently bright pixels among the divided subblocks.

A control unit 118 is provided for performing local histogram expansion to improve contrast of the sub block selected by the selection unit 116. At this time, the controller 118 does not target all pixels located in the sub-block, and performs histogram expansion only on pixels that satisfy the condition of Equation (4). Therefore, according to the present embodiment, when the histogram expansion is performed, the dark area becomes darker. The reason for doing so is to limit and implement the local histogram extension technique.

--------- 식(4)

--------- Formula (4)

는 각 서브블록의 입력 픽셀의 휘도 값,

는 블록별 휘도 평균값, σ_b는 표준편차이다. here,

Is the luminance value of the input pixel of each subblock,

Is the average luminance value for each block, and sigma _b is the standard deviation.

The controller 118 performs the histogram expansion using the algorithm of Equation (5).

-------- 식(5)

-------- Formula (5)

는 히스토그램 확장된 출력 픽셀의 휘도 값,

는 각 서브블록의 입력 픽셀의 휘도 값, α는 확장계수(즉 블록 노이즈를 제한할 수 있는 범위 내에서 선택할 수 있는 값),

는 각 블록별 휘도 평균값, σ_b는 표준편차 정보이다. here,

Is the luminance value of the histogram extended output pixel,

Is the luminance value of the input pixel of each subblock, α is the expansion coefficient (that is, a value selectable within a range that can limit block noise),

Is the average luminance value of each block, and sigma _b is standard deviation information.

After the histogram expansion of the selected subblock is performed, the backlight controller 120 adjusts a supply current to an LED which is a light source of the pixel based on the information on each subblock. The backlight controller 120 also integrates the Y-Cb-Cr image divided into the sub-blocks into one, and converts the image into an RGB image.

Next, a control method of the image contrast control device having the above configuration will be described with reference to FIG. 2. 2 is a flowchart illustrating a method of controlling image contrast for a local dimming backlight according to a preferred embodiment of the present invention.

First, the converting unit 102 converts the input RGB image into a Y-Cb-Cr image using Equations (1) to (3) (s100).

)을 추출한다(s102). When converted into the YCbCr image, the first extractor 104 extracts a luminance (Y) component from the Y-Cb-Cr image, and the second extractor 106 extracts the Y-Cb-Cr image. Overall luminance average value

) Is extracted (s102).

)이 추출된 다음에는, 본 실시 예의 히스토그램 확장을 적용하기 위해 상기 분할부(108)가 상기 Y-Cb-Cr 영상을 M * N 개의 서브 블록 크기로 분할한다(s104). 상기 분할은 상기 Y-Cb-Cr 영상의 전체 크기 및 해상도에 따라 다양한 개수로 분할된다.The overall luminance average value (

) Is extracted, the divider 108 divides the Y-Cb-Cr image into M * N sub-block sizes in order to apply the histogram extension of the present embodiment (s104). The division is divided into various numbers according to the overall size and resolution of the Y-Cb-Cr image.

) 을 계산하고, 아울러 제 2 계산부(112)는 표준편차(σ_b)정 보를 계산한다(s106). 상기 표준편차(σ_b)는 히스토그램 확장 정도를 제어하기 위한 값으로 제공된다. The first calculator 110 calculates a luminance average value of each block for the sub-blocks divided into M * N pieces.

), And the second calculation unit 112 calculates standard deviation (σ _b ) information (s106). The standard deviation σ _b is provided as a value for controlling the degree of histogram extension.

)과 블록별 휘도 평균값(

)이 계산되면, 상기 비교부(114)는 상기 서브 블록별로 전체 휘도 평균값(

)과 블록별 휘도 평균값(

)을 비교한다(s108). 이는 히스토그램 확장을 수행할 서브 블록을 선택하기 위함이다. The overall luminance average value (

) And block-by-block luminance average (

Is calculated, the comparator 114 calculates an overall luminance average value for each of the sub-blocks.

) And block-by-block luminance average (

) Is compared (s108). This is to select a subblock to perform histogram expansion.

<

을 만족하는 서브 블록으로 한정된다. 만약 상기 비교결과에 따라 상기

<

을 만족하는 서브 블록이 존재하지 않는다면, 아래의 제 118 단계로 진행하여 분할된 영상을 통합하고 출력하게 된다. According to the comparison result, the selector 116 selects a subblock to which the histogram expansion technique is applied from the divided subblocks (S110). The sub-block selection is

<

It is limited to a sub block that satisfies. If according to the comparison result

<

If there is no sub-block satisfying the, the process proceeds to step 118 below to integrate and output the divided image.

However, when a sub block to which the histogram extension technique is to be applied is selected by the selection unit 116, the controller 118 performs an independent histogram extension control process in units of blocks of the selected sub block to realize contrast enhancement.

)의 표준편차 내로 한정하여 분할 이전의 원 영상의 전체 휘도 평균값(

) 이하의 값을 가지는 픽셀 영역을 결정한다. 상기 히스토그램 확장이 적용될 픽셀 영역의 결정 예가 도 3에 도시하고 있다. In this case, the control unit 118 does not perform histogram expansion on all pixels located in the selected subblock, and uses Equation (4) above to determine the pixel range to which the actual histogram extension is applied among the pixels located in the selected subblock. Determine using (s112). That is, the pixel range is a luminance average value for each block (

Total luminance mean value of the original image before dividing (limited to the standard deviation of

A pixel area having the following value is determined. An example of determining a pixel area to which the histogram extension is applied is shown in FIG. 3.

)에서 많이 벗어난 픽셀 값에 대해서는 히스토그램 확장 폭이 좁게 되게 하고, 반면 상기 블록별 휘도 평균값(

)에 근접된 픽셀 값에 대해서는 히스토그램 확장 폭이 넓게 되도록 제어한다(s114). 즉 상기 히스토그램 확장 정도는 각 서브 블록의 표준편차(σ_b)가 작을수록 그리고 서브 블록의 휘도 평균값(

)이 클수록 증가한다. When the pixel region in which the histogram expansion is to be performed is determined, the controller 118 applies the inverse of the standard deviation σ _b to the determined pixel region as shown in Equation (5).

For pixel values that deviate significantly from), the histogram extension width is narrowed, whereas the luminance average value of each block (

For the pixel value close to), it is controlled to widen the histogram extension (S114). That is, the histogram expansion degree is obtained by decreasing the standard deviation (σ _b ) of each sub block and by the luminance average value of the sub block (

Increases as) increases.

4 illustrates an example in which the histogram expansion is performed by the above method.

FIG. 4A is a Spiderman original image screen with a 24-bit RGB image and a resolution of 600 * 285.

<

을 만족하는 5개의 서브 블록(흰색 점선)을 구하고, 그 서브 블록에서 원 영상의 전체 휘도 평균값(

) 이하의 값을 가지는 픽셀에만 표준편차(σ_b)를 '10'으로 하여 히스토그램 확장을 수행하였다. 그 결과 화면이 도 4b에 도시되어 있다. 참고로, 도 4c는 도 4a의 원 형상화면을 종래 히스토그램 평활화를 적용하여 얻어진 결과 화면이다.And dividing the circular screen into nine subblock regions of 3 * 3 in total, and then

<

Five subblocks (dotted white lines) satisfying are obtained, and the total luminance average value of the original image is obtained from the subblock.

Histogram expansion was performed with the standard deviation σ _b as '10' only for pixels with values less than or equal to). The resulting screen is shown in FIG. 4B. For reference, FIG. 4C is a result screen obtained by applying the conventional histogram smoothing to the circular screen of FIG. 4A.

Thus, it can be seen that there is a significant difference when comparing the screen of FIG. 4B with the screen of FIG. 4C. That is, in FIG. 4C, histogram smoothing is applied to all pixels of the original image by using the same function, resulting in a whitening of the color as a whole. On the other hand, as shown in the screen according to the embodiment of FIG. 4B, the selected five sub-blocks are regions in which pixel values constituting the pixels have relatively dark values, and the histogram expansion is performed within a limited range. In comparison, the darker areas become darker.

A graph comparing the histograms of FIGS. 4A and 4B is shown in FIG. 5. Referring to FIG. 5, when the histogram (dotted line) of the original image is compared with the case where the present embodiment is applied (solid line), the pixels having the luminance average value of '54' level or less of the original image are mapped near the '0' level, and thus, the overall dynamic range. You can see that this is enlarged. At this time, it can be seen that the extent of expansion of the histogram is proportional to the expansion coefficient α, which is shown in FIG. 6. 6 is a graph showing a change state of the histogram according to the expansion coefficient α.

On the other hand, after the histogram expansion is performed for only some pixels in the divided sub-blocks as described above, the backlight controller 120 receives the current supplied to the LED which is a light source representing the pixel based on the information of the sub-blocks. Control (s116). After integrating the divided image (s118), the image is converted into an RGB image and output (s120). The output image is the same as FIG. 4B described above.

On the other hand, the degree of contrast improvement through the present embodiment can be confirmed by the method proposed by C.F.Hsu et al. [1]. This is shown in the following [Table 1]. [Table 1] shows the ratios of the luminance component values 'C25' and 'C75' corresponding to 25% and 75% on the histogram of the original image and the improved image.

TABLE 1

division C25 C75 C75 / C25 Video Improvement Cost Original video 24 65 2.7 0 α = 5 22 66 3 11% α = 10 20 67 3.4 24% α = 20 18 69 3.8 41%

As shown in Table 1, it can be seen that the image improvement ratio is proportionally improved according to the expansion coefficient α. In particular, when the expansion coefficient (α) is 20, an improvement of 41% can be confirmed.

As described above, the present invention divides the input image into blocks of the same size and analyzes pixel information in each sub-block to generate histograms, averages, and standard deviations of brightness components (Y), and then blocks. It can be seen that the contrast is improved by performing the histogram expansion independently for each predetermined pixel.

Although described with reference to the illustrated embodiment of the present invention as described above, this is merely exemplary, those skilled in the art to which the present invention pertains various modifications without departing from the spirit and scope of the present invention. It will be apparent that other embodiments may be modified and equivalent. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1 is a block diagram of an image contrast control apparatus according to a preferred embodiment of the present invention

2 is a flowchart illustrating a method of controlling image contrast for local dimming backlight according to an exemplary embodiment of the present invention.

3 is an exemplary diagram in which a pixel area to which histogram extension is applied is determined in FIG. 2;

4A to 4C are diagrams illustrating a histogram expansion performed according to an embodiment of the present invention.

5 is a graph comparing the histograms of FIGS. 4A and 4B.

6 is a graph showing a change state of the histogram according to the expansion coefficient α.

* Description of the symbols for the main parts of the drawings *

102: converting unit 104: first extracting unit

106: second extraction unit 108: division

110: first calculation unit 112: second calculation unit

114: comparison unit 116: selection unit

118 control unit 120 backlight control unit

Claims (8)

Converting an RGB image to a Y-Cb-Cr image; Extracting a luminance (Y) component from the Y-Cb-Cr image; Overall luminance average value of the Y-Cb-Cr image

Calculating; The overall luminance average value (

Is calculated, the size of the Y-Cb-Cr image is divided into M * N sub-block sizes; The luminance average value of each block for each of the divided subblocks (

) And the standard deviation information σ _b ; The luminance average value of each block among the divided subblocks (

Is the overall luminance average value (

Selecting only subblocks smaller than); And The luminance average value of each block among the selected subblocks

The overall luminance average value (limited to the standard deviation σ _b ) of

And applying local histogram stretching only to pixels having a value less than or equal to the following. The method of claim 1, Controlling a current supplied to a light source of the pixel based on information of each sub block in which the histogram expansion is performed; And And integrating the image divided into the sub-blocks, converting the image into an RGB image, and outputting the converted RGB image. The method of claim 1 or 2, wherein the local histogram expansion technique, The degree of expansion of the histogram is controlled in consideration of the distribution of pixels located in the sub-block, wherein the standard deviation of the sub-block is small and the luminance average value of each block of the sub-block (

And) increasing the histogram extension to increase the image contrast control method for the local dimming backlight. The method of claim 3, wherein And the local histogram extension technique makes only darker regions darker than the original image with respect to some pixel values of the selected subblocks. The method according to claim 1 or 2, And the sub-block unselected in the selecting step maintains the luminance value of the original image as it is. A first extracting unit which extracts a luminance (Y) component of Y-Cb-Cr image information converted from an RGB image; Luminance average value of the Y-Cb-Cr image information (

A second extracting unit for extracting; A divider dividing the size of the Y-Cb-Cr image information into M * N sub-block sizes; The luminance average value of each block for each of the divided subblocks (

A first calculation unit for obtaining _&quot; ) and a second calculation unit for obtaining standard deviation information [sigma] _b ; Luminance average value of the Y-Cb-Cr image information (

) And the luminance average value of each block (

Comparing unit for comparing the size of; A selection unit for selecting a sub-block to which the histogram expansion technique is applied among the divided blocks according to the comparison result; And Image contrast control for local dimming backlight configured to include a control unit that performs local histogram stretching by only the pixels corresponding to the following equation (1) among the selected subblocks by the following equation (2) Device.

-σ _b <

<

+ σ _b ----- Equation (1)

----- Formula (2) here,

Is the luminance value of the histogram extended output pixel,

Is the luminance value of the input pixel of each subblock, α is the expansion coefficient,

Is the average luminance value for each block, and σ _b is the standard deviation information. The method of claim 6, The selection unit

<

The image contrast control device for local dimming backlight, characterized in that for selecting a sub block that satisfies the sub block to apply the histogram expansion technique. The method of claim 6, And a backlight controller configured to control a supply current supplied to a light source of the pixel based on the information of each sub-block in which the histogram expansion is performed.

Images