KR20020067852A - Image Quality Control Method of Flat Panel Display - Google Patents

Abstract

The present invention relates to a method for adjusting the image quality of a flat panel display panel to enable image quality improvement.

According to an exemplary embodiment of the present invention, a method for adjusting image quality of a flat panel display panel compares a luminance value of input image data in a screen in which a histogram indicating contrast is expanded with a predetermined reference luminance value, and as a result, the luminance value of the input image data is a reference luminance. Adding the luminance value of the image data that is different from the value with a predetermined weight, dividing the screen into at least two blocks, calculating a total sum of luminance of each block, and selecting the block having the largest value. Determining a relative luminance value of the blocks as a reference, adjusting a weight corresponding to each of the blocks according to the relative luminance value, and generating a signal for controlling the backlight according to the relative luminance value.

According to the present invention, the image quality can be improved by obtaining an optimal image by selectively varying the weight slope for each section according to the image type, and in order to calculate the weight value, a mode detector and a barrel shifter By using, only the difference between the maximum pixel value and the minimum pixel value in one image reduces the relative amount of computation, thereby preventing underfloor or overflow.

Description

Image Quality Control Method of Flat Panel Display

The present invention relates to a method for adjusting the image quality of a flat panel display panel to enable image quality improvement.

In high information society, research on display device technology for displaying information is being actively conducted. Information transmission will be mainstream of a system that simultaneously sends and receives text, voice and video. The basic functions of the display device required by the system should have the characteristics of human-centered, environmental-oriented, advanced, and highly functionalized. In this era, research on flat panel display devices is actively underway in Korea. Display devices are classified into a cathode ray tube device (hereinafter referred to as "CRT") and a flat panel display device (hereinafter referred to as "FPD"). Currently, most of the research results of liquid crystal display devices (Liquid Crystal Display; "LCD") in the field of flat panel display are forming most of the results, and Plasma Display Panel (hereinafter referred to as "PDP") is easy to large screen The research on this is also actively conducted by showing the quality of CRT level.

FPD has lower power consumption, higher resolution, thinner and lighter weight than CRT and adopts digital driving method. The advantages of this digital processing method are noise, and increasing the number of bits widens the range of numerical values that can be expressed, enabling arithmetic processing or accumulation with low degradation and high image quality. Therefore, it is necessary to study the digital signal processing technique for the image displayed on the FPD. Among them, image processing methods for improving image quality include contrast control, noise reduction, edge restoration, edge enhancement, gamma correction, and the like. Contrast adjustment, one of these areas, can improve image quality without increasing information. The most widely used method to improve image quality is to use a histogram distribution. A histogram is a summary of the brightness of a video. That is, when light and dark points are distributed, the range and value of the distribution are expressed. When digitizing an image, use brightness as wide as possible to get good quality.

Contrast ranges from the darkest to the brightest areas of the image. Images with high contrast range from dark to bright areas. Therefore, as the contrast of the image increases, the image can be viewed in more detail, which is due to the pure visual action in which the amount of information of the image does not increase at all.

Human visual action is more sensitive to the contrast of luminosity, and the distribution of such contrast can be seen by histogram. The histogram of the image provides information for determining the contrast value of the image, and also provides information on the contrast and contrast distribution of the contrast constituting the image. The histogram of the dark image is biased to the left of the pixel value distribution, and the bright image of the dark image is equal to the histogram of the h value of the pixel value distribution to the right. If the image is ideal, the histogram of the image is uniformly distributed.

Methods used in the digital image processing part include a look-up table, histogram sliding, and histogram stretching.

Looking at the contrast adjustment method by the above three methods are as follows.

1. Lookup Table Method

In the lookup table method, the current pixel value becomes the address portion of the lookup table, and the contents of the lookup table address are converted into new pixel values and output. This method is represented by Equation 1, and shows an image using a 3-bit lookup table in FIG.

New pixel = DATA [Add (Input Pixel)]

[Where, Add (*): * Address of address]

The new pixel value in FIG. 1 is the array data indicated by the index of the array, and the generation of the new pixel value is generated by performing a series of operations indicated by this index for each pixel. This scheme complicates the circuit structure because the circuit for the lookup table calculation and the separate memory are required to store the calculated value, and the separate circuit is required to prevent the under / overflow of the processed data. Has its drawbacks.

2. Histogram sliding method

In the histogram sliding method, an output pixel value is determined by giving a predetermined weight to an input pixel value, and can be expressed by Equation 2.

New Pixel = Input Pixel × Weight

Figure 2 shows the change in histogram distribution in this way.

The gray straight line in FIG. 2A shows the function of the input / output relationship. This functional relationship extends the area of the input pixel into a uniform area according to the slope of the straight line determined by the distribution area of the histogram. That is, the area of the input pixels 63 to 127 is expanded to the area of the input pixels from 0 to 255.

The difference between this method and the lookup table method is that it does not require memory to store the lookup table value and the weight is constant. However, since the distribution of the histogram is not constant, a circuit for weight calculation is additionally required to prevent under / overflow.

3. Contrast stretching method

This method, which is most commonly used, is effectively applied to certain parts, i.e., images with a contrast value in the center. The output pixel value is represented by equation (3).

New Pixel = [(Input pixel-Low pixel) / (High pixel-Low pixel)] × 255

As shown in Equation 3, when the subtraction operation is performed using the pixel value having the lowest value in the image and the subtraction operation is performed, the histogram is shifted to the left. Extended histogram distribution to cover the entire region. Although this method does not generate under / overflow, it must be composed of a multiplier and an accumulator with a complicated structure, and thus it is difficult to apply in a part requiring real-time processing such as an LCD driving system.

Accordingly, an object of the present invention is to provide a method for adjusting the image quality of a flat panel display panel, which is capable of improving image quality by selectively changing the weight gradient according to the image type and adjusting the brightness of the backlight according to the image type.

1 is a diagram illustrating an operation using a conventional lookup table;

Figure 2 is a distribution diagram of the histogram by the conventional histogram sliding method.

3 is a block diagram illustrating a method for adjusting image quality according to the present invention;

4 is a view for explaining a threshold for image type classification according to the present invention;

5 is a diagram illustrating a weight for an image type according to the present invention.

6 shows simulation results of a dark image type according to the present invention.

7 shows simulation results of an image type of an intermediate erection according to the present invention.

8 shows a simulation result of a bright image type according to the invention.

9 is a diagram showing a simulation result in a grayscale image according to the present invention;

10 to 13 is a view showing a simulation result by the weight according to the present invention.

14 is a diagram for explaining an output signal of a backlight control signal according to the present invention;

Fig. 15 illustrates the effect according to the present invention.

In order to achieve the above object, the image quality adjusting method of the flat panel display panel according to the present invention compares the luminance value of the input image data in the screen to expand the histogram indicating the contrast with a predetermined reference luminance value and as a result the input image Adding the luminance value of the image data whose luminance value differs from the reference luminance value by a predetermined weight, dividing the screen into at least two blocks, calculating a total sum of luminance of each block, Determining a relative luminance value of the blocks based on the block having the largest value, adjusting a weight corresponding to each of the blocks according to the relative luminance value, and a signal for controlling the backlight according to the relative luminance value Generating steps.

Other objects and features of the present invention in addition to the above object will be apparent from the description of the embodiments with reference to the accompanying drawings.

Hereinafter, exemplary embodiments of the present invention will be described with reference to FIGS. 3 to 13.

3 is a block diagram illustrating a method of adjusting image quality of a flat panel display device according to the present invention.

The upper end of FIG. 3 searches for the maximum and minimum pixel values when the input pixel value is input, and determines the weight of the retrieved pixel value. The area of the histogram is expanded according to the determined value of the weight.

When the input pixel value is input, the lower part searches for the degree of contrast of Dark, Middle, Bright, Dark & Bright, and determines the inclination according to the degree of contrast. According to this inclination determination, the inclination of each area is changed.

Thereafter, the output pixel value is determined according to the slope of the area of the extended histogram of the upper part and the degree of contrast of the lower part.

In addition, when the degree of contrast is recognized according to the input pixel value of the lower part, a signal for controlling the backlight is generated using the backlight lamp controller.

In detail, first, the expansion of the histogram allows the weight of the contrast stretching scheme to be adjusted by a user selector (hereinafter referred to as “US”).

New pixel = (Input Pixel-MIN Pixel) × (M + US)

M (Mltiple): INT (255 / {MAX pixel-MIN pixel})

If (16≤DR≤127) then US = {-2, -1,0,1,2}

else US = {0, 0.125, 0.25, 0.375, 0.5, 0.625, 0.75}

As shown in Equation 4, since the subtraction operation is performed on the pixel value currently input using the smallest pixel value, the smallest value among the pixel values of the image becomes 0 and an underfloor can be prevented. We also compute the weight (M + US) to extend the histogram to cover the entire area without overflow. The user may gradually adjust the contrast of the image by using a user selector (US). Here, if the difference ("DR") exists between 16 and 127, the reason for the US value of -2 to 2 is to extend the histogram to the area where no overflow occurs. If more than, multiply by less than 2 weights to expand.

The M (Mltiple) value is determined by Equation 5, but does not calculate the M value using an accumulator, but uses bit information of a value corresponding to the difference between the maximum value and the minimum value of the histogram. Therefore, a simple hardware configuration is possible.

Histogram Weight Binary Difference range M M + US 0001XXXX 16 to 31 8 6, 7, 8, 9, 10 001XXXXX 32 to 63 4 2, 3, 4, 5, 6 01XXXXXX 64-127 2 1, 2, 3, 4 1XXXXXXX 128-225 One 1.125, 1.25, 1.375, 1.5, 1.625, 1.75

(X: don't care)

In the mode detection part, the image type is determined. That is, the input image is classified into four types (Dark, Middle, Bright, Dark & Bright). As shown in FIG. 4, the pixel value distribution section is set in three stages to determine the image type by determining a section having the largest value by calculating the sum of pixel values in each section within a frame of the input image.

As shown in FIG. 4, when the pixel values are most present within 0 to 64, the dark image is determined, and when it is between 64 and 192, the bright image is determined. Also, if the pixel values exist from 0 to 64 and 192 can not be ignored, the dark and bright image is determined.

The weight of the pixel value is determined for each section after the histogram stretching according to the determined image type. This process can be performed in real time by applying the pixel value to the address of the ROM and generating a new pixel value weighted by the data value contained in the ROM, as in the conventional lookup table method. 5 is a graph of weights assigned to respective image types.

Through the weight graph shown in FIG. 5, the slope for each image type may be determined differently.

These can be represented as follows by analyzing the performance of the method for the present invention by various experimental images.

Lena image and Salesman's image were used as general images, and gray scale images were also simulated.

6 to 8 are simulation results of the image improved by the contrast adjustment method for each type of Lena image. FIG. 6 is a dark image, FIG. 7 is an image of medium brightness, and FIG. 8 is a light image type. It is expressed.

B in each figure is defined as stretching 1, which is the result when the histogram distribution is expanded to maximum before overflow occurs before weighting. In addition, c in each drawing is set to stretch 2, which is a result of the image given by the weight for each section in the expanded image.

9 is a simulation result after applying each method in the grayscale image. FIG. 9A illustrates an original grayscale image, FIG. 9B illustrates a dark image, FIG. 9C illustrates an image of medium brightness, and FIG. 9D illustrates a grayscale image of a bright image type.

10 to 13, as a result of simulating each contrast image through respective weights, it can be seen that the smaller the difference between the maximum value and the minimum value of the pixel, the sharper the contrast of the image.

FIG. 10 is a case of an original image, in which FIG. 10A shows contrast when the maximum value and the minimum value are (230, 22), FIG. 10B is (100, 20), and FIG. 10C is (230, 162).

FIG. 11 is a dark image, and FIG. 11A shows contrast when the maximum value and the minimum value are (228, 0), FIG. 11B is (248, 20), and FIG. 11C is (230, 162).

FIG. 12 is an image of medium brightness, in which FIG. 12A shows contrast when the maximum and minimum values are (239, 22), FIG. 12B shows (251, 0), and FIG. 12C shows (255, 0). will be.

FIG. 13 is a case of a bright image. FIG. 13A shows contrast when the maximum value and the minimum value are (239, 22), FIG. 13B is (251, 0), and FIG. 13C is (255, 0).

In FIG. 3, when Dark, Middle, Bright, Dark & Bright are determined based on the result of the mode determination, the data is applied to the backlight controller, which is a backlight control signal PWM having a wide pulse width when the determined mode is Bright Mode. 14a is selected and outputted, and in dark mode, the pulse width narrow backlight control signal 14c is outputted. In other modes, Fig. 14B, which is the control signal PWM of the intermediate pulse width, is output.

This is to adaptively select and apply a voltage applied to the backlight according to the determined mode. In this way, a contrast synergistic effect can be expected. 15 is a view for explaining the effect of improving contrast and adjusting the brightness of the backlight according to the brightness state of the screen according to the present invention. As shown in FIG. 14, the brightness is improved and the contrast is improved in the bright screen. In the dark screen, the contrast is improved but the brightness is reduced, thereby maximizing the contrast.

As described above, in the contrast adjustment method of the flat panel display apparatus according to the present invention, the image quality can be improved by obtaining an optimal image by selectively varying the weight slope for each section according to the image type, and accumulating to calculate the weight value. By using a mode detector and a barrel shifter without using a multiplier, it uses the difference between the maximum pixel value and the minimum pixel value in only one image. By using MD and BS to calculate the weighted value, the difference between the maximum pixel value and the minimum pixel value in a single image is reduced by using the MD and BS. Can be. In addition, the brightness of the backlight is adaptively changed according to the image type, thereby providing an advantage of maximizing contrast synergy.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the technical spirit of the present invention. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification but should be defined by the claims.

Claims (4)

In the image processing method using the histogram distribution in order to improve the image quality of the flat panel display device, The luminance value of the input image data in the screen in which the histogram indicating the contrast is expanded is compared with a predetermined reference luminance value, and as a result, the luminance value of the image data in which the luminance value of the input image data differs from the reference luminance value. Adding to a predetermined weight, Dividing the screen into at least two blocks; Calculating a sum of luminance of each of the blocks and determining relative luminance values of the blocks based on the block having the largest value; Adjusting the weight corresponding to each of the blocks according to the relative luminance value; And generating a signal for controlling a backlight according to the relative luminance value. The method of claim 1, The adding of the weights may include processing the lowest luminance value of the input image data as 0; And expanding the histogram by adding a weight to the histogram shifted to the left according to the result so that the maximum luminance value covers the entire area. The method of claim 1, And a block of the screen is composed of a light image, a dark image, and an image of medium brightness according to a relative luminance value. The method of claim 1, And varying a pulse width of a signal for controlling the backlight.