CN104867453B - The dot structure conversion method and its device of display screen - Google Patents

Abstract

The invention discloses the dot structure conversion methods and its device of a kind of display screen.The dot structure conversion method includes the following steps：Take out the three primary colours grayscale of all pixels unit in the image of the RGB forms of input；The three primary colours grayscale for taking out adjacent three to five pixel units in original image one by one, is converted to new grayscale so that the grayscale difference of two squares between original pixel unit and new pixel unit is minimum；And after obtaining all new grayscale, display is re-started to the image inputted in a manner of RGBG.Thus existing RGB being shown to, being converted to RGBG shows, solves the mismatch problem of dot structure.

Description

Pixel structure conversion method and device of display screen

Technical Field

The invention relates to the field of display, in particular to a pixel structure conversion technology of a display screen.

Background

Compared with other display screens (such as an LCD, etc.), an AMOLED (Active-matrix organic light-emitting diode) screen has many advantages, including an omni-directional 180-degree viewing angle, a fast response speed, and convenience for an ultra-thin design. The AMOLED display technology is widely valued internationally, and is actively developed by international large enterprises represented by samsung, LG, sharp, and friendship. Currently, the AMOLED display screen is mainly applied to a mobile phone and a portable media player. Based on the data analysis provided by ABI, the shipment of smart phones below $ 200 in 2013 will grow to about 2.38 million stations, while the number will proliferate to 7.58 million stations in 2018. The displaysearch data shows that in recent years, the proportion of AMOLED panels used globally in the field of mobile phones to the total output volume has rapidly risen, driven by the rapid growth of the smartphone market, and will exceed 70% by 2016. According to the data of display research, the global shipment of AMOLED panels reaches 3.95 hundred million pieces by 2016, wherein about 2.83 hundred million pieces are used for mobile phones. Currently, in the field of high-end electronic products, the AMOLED display screen has gradually become one of the important differentiation factors between products. In addition to the promotion of the smart phone market, the yield can be gradually improved along with the further maturity of the technology in the future, the production cost is greatly reduced, and the defect of high product price is gradually relieved. In addition, the future large-size will be produced in large scale, and the application of the AMOLED panel in the fields of notebook computers, netbooks, displays and televisions will be increased. According to the displaysearch data, it is expected that the growth will be substantially $ 71 billion by 2016, during which the annual average composite growth rate reaches 36%.

Fig. 2 shows two AMOLED pixel structures, where the pixel structure shown in fig. 2(a) is RGBW and the pixel structure shown in fig. 2(b) is RGBG, where green sub-pixels are alternately arranged between red and blue sub-pixels.

As shown in fig. 2(b), for an RGBG panel, each pixel is composed of two sub-pixels: RG or BG. As is well known, in order to display all colors in the CIE chromaticity diagram, each pixel should include three primary color sub-pixels RGB, so that a pixel borrowing method is adopted to display all colors in a pixel unit including only two sub-pixels. For example, in an RGBG pixel structure, a pixel cell composed of RG needs to borrow the missing B sub-pixel from the neighboring pixel cell to compose the RGB three primary colors. Compared to a conventional RGB stripe pixel structure, the number of sub-pixels required for an RGBG screen is one third less than that required for an RGB stripe structure for the same resolution, which greatly reduces the cost. Furthermore, RGBG screens can achieve a resolution 1.5 times that of conventional stripe RGB for the same physical size of the screen. Finally, for the same physical size and the same resolution, the sub-pixel area of the RGBG screen can be made larger, and when the same brightness is displayed, the required current is smaller, which is beneficial to slowing down the aging of the AMOLED.

In summary, most of the image pixels are arranged in RGB stripes, and in order to display images on RGBG panels, a method for converting the RGB display into the RGBG display is urgently needed in the art.

Disclosure of Invention

The invention aims to provide a pixel structure conversion method and a pixel structure conversion device for a display screen, which are used for converting the existing RGB display into RGBG display based on a gray level square error and solving the problem of mismatching of pixel structures.

In order to solve the above technical problem, an embodiment of the present invention discloses a method for converting a pixel structure of a display screen, including the following steps:

taking out three primary color gray scales of all pixel units in the input RGB type image;

three primary color gray scales of three to five adjacent pixel units of the input image are taken out one by one and are converted into new gray scales, so that the gray scale square error between the original pixel unit and the new pixel unit is minimum; and

after all new gray scales are obtained, the input image is redisplayed in the RGBG manner.

The embodiment of the invention also discloses a pixel structure conversion device of a display screen, which comprises:

a gray scale obtaining part for taking out three primary color gray scales of all pixel units in the input RGB type image;

the gray scale conversion part is used for taking out the three primary color gray scales of the three to five adjacent pixel units one by one and converting the three primary color gray scales into new gray scales so as to ensure that the gray scale square difference between the original pixel unit and the new pixel unit is minimum; and

and a display unit for displaying the input image again in the RGBG mode after all new gray scales are obtained.

Compared with the prior art, the implementation mode of the invention has the main differences and the effects that:

by regenerating a new gray scale for the gray scale of the corresponding pixel unit of the original image, the gray scale square difference between the original pixel unit (original image in RGB format) and the new pixel unit (RGBG format) is reduced to the minimum, and the color edge effect of the RGBG format image can be reduced.

Further, the conversion into the new gray level includes the steps of: and constructing a function of which the gray scale square difference changes along with the gray scales of red and blue channels in the display screen, solving the corresponding gray scale when the gray scale square error is minimum, and simultaneously keeping the green gray scale in the display screen unchanged, so that the display effect generated by the obtained new gray scale is better.

Further, the display screen is an AMOLED screen, although the invention is applicable to all display screens with RGBG pixel structures, and is not limited to the AMOLED screen.

Drawings

Fig. 1 is a flowchart illustrating a pixel structure conversion method according to a first embodiment of the invention.

Fig. 2 is a schematic diagram of two pixel structures of the display panel.

Fig. 3 is a schematic diagram of a RGB to RGBG pixel structure conversion.

FIG. 4 is a schematic diagram of determining gray level squared differences in the pixel structure conversion method of FIG. 1.

Fig. 5 is a schematic structural diagram of a pixel structure conversion device according to a second embodiment of the invention.

Detailed Description

In the following description, numerous technical details are set forth in order to provide a better understanding of the present application. However, it will be understood by those skilled in the art that the technical solutions claimed in the present application can be implemented without these technical details and with various changes and modifications based on the following embodiments.

First, prior to the detailed description of the invention, some basic knowledge about the conversion of the pixel structure in the present invention will be described.

Through direct mapping, the gray scale original value of each sub-pixel in the original image can be directly assigned to the sub-pixel at the corresponding position in the RGBG screen. As shown in FIG. 3, R of pixel 1 in the original image is set_2iAnd G_2iRespectively assigned to R of pixel 1' in RGBG screen_2i' and G_2i' of the original image, B of the pixel 2_2i+1And G_2i+1B respectively assigned to pixels 2' in RGBG screens_2i+1' and G_2i+1'。

A schematic diagram of the RGB to RGBG pixel structure conversion is shown in fig. 3. Fig. 3(a) shows 4 consecutive pixel elements in an RGB format image, and fig. 3(b) shows 4 consecutive pixel elements in corresponding positions in an RGBG panel. The pixel structure conversion from RGB to RGBG is to obtain the gray scale of the pixel unit at the corresponding position in the RGBG screen based on the known gray scale of each pixel unit in the original image. Since the minimum repeating unit of the RGBG panel occupies 2 pixels, the number of the pixels is 2i,2i +1, where i is 0,1,2, …, N/2-1, and N is the total number of pixels of the image. In fig. 3(a), pixel 1 and pixel 2 inside the large dashed line frame correspond to pixel 1 'and pixel 2' inside the large dashed line frame in fig. 3(b), respectively.

From the viewpoint of data processing, it is necessary to store the gray scales of each pixel unit in the original image in a memory, read the gray scales pixel by pixel, process the gray scales according to a specific conversion method, and store the processed gray scales pixel by pixel in another storage area of the memory.

As is well known, in the CIE (international commission on illumination) chromaticity diagram, all colors can be represented by a combination of three primary colors of red, green and blue. In an RGBG panel, however, each pixel cell contains only 2 sub-pixels, i.e. only 2 primaries. Thus, each pixel cell in an RGBG panel needs to borrow the missing sub-pixel from the neighboring pixel cell in order to display all colors. As shown in fig. 3(b), a pixel cell in an RGBG type panel includes 2 sub-pixels, RG or BG. As described above, in order to be able to display all colors in the CIE chromaticity diagram, one pixel unit should include three primary color sub-pixels. Since each pixel cell of an RGBG panel is composed of two sub-pixels (RG or BG), the missing sub-pixel needs to be borrowed from the neighboring pixels.

As shown in FIG. 3(b), two sub-pixels of the pixel 1' are R_2i' and G_2i', the pixel cell lacks the blue sub-pixel, and therefore, borrows B from the neighboring pixel 2_2i+1', to form the complete three primary colors.

The visual effect evaluation index of RGBG image display is briefly described below.

Fig. 3(a) shows an original image in RGB format, and fig. 3(b) shows the corresponding RGBG. Taking pixel 1 as an example, pixel 1 in the original image corresponds to 1 'in RGBG, and the pixel unit only includes 2 sub-pixels (R2i' and G2i '), and the blue sub-pixel with the nearest distance needs to be borrowed from the adjacent pixels, where B in pixel 2' is borrowed_2i+1', the 3 sub-pixels constitute one pixel 1 ", which is referred to as a virtual pixel of the pixel 1'.

Original pixel:

pixel 1 (R)_2i,G_2i,B_2i)

Pixel 2 (R)_2i+1,G_2i+1,B_2i+1)

Virtual pixels:

pixel 1 ″ (R)_2i',G_2i',B_2i+1') wherein B is_2i+1'is borrowed from the pixel 2'.

Pixel 2 "(R)_2i+2',G_2i+1',B_2i+1') wherein R is_2i+2'is borrowed from pixel 3'.

The image composed of all the virtual pixels is a virtual image. The virtual pixel is formed by the principle that sub-pixels in RGBG are borrowed by the pixel. Since the three primary color gray scale of the virtual pixel determines the color and brightness of the corresponding pixel unit, the quality of the virtual image can reflect whether the conversion is superior or not. For the same image, the number of sub-pixels used in the RGBG type panel is reduced by one third compared with the number of sub-pixels of the original image. Since the display screen of this format uses fewer sub-pixels to display the same image, the visual effect of the display on the screen side of the image is more or less reduced. It is necessary to evaluate the visual effect of the image displayed on the RGBG screen. As shown in fig. 3, the color and brightness of pixel 1 and pixel 2 are determined by the gray scale of their respective pixels. Accordingly, the color and brightness of the pixel 1 'and the pixel 2' are determined by the gray scale size of their respective dummy pixels, respectively. Therefore, the difference between the original pixel cell and the virtual pixel cell should be used to evaluate the visual effect of the image displayed on the RGBG screen.

Gray scale squared error

The gray-scale square error is a common visual effect evaluation index, as shown in fig. 4, the upper half of fig. 4 is 4 pixels in an RGB format image, and the lower half is 4 pixels in corresponding positions in an RGBG screen. Since the minimum repeating unit of the RGBG panel occupies 2 pixels, the pixel is indexed by 2i,2i +1, where i ═ 0,1,2, …, N/2-1, and N is the total number of pixels in the image. For the red channel, the gray scale squared error is:

wherein N is the total pixel number in the image, R'_2iAnd R'_2i+2Is the unknown to be solved, F_RIs with R'_2iAnd R'_2i+2Function of variation, R_2i、R_2i+1The gray scale of the original image is known. As can be seen from FIG. 4, the sub-pixel R of the image_2iAnd sub-pixel R 'of display screen'_2iThere is a one-to-one correspondence in spatial position, so the two are compared. However, for the sub-pixel R of the image_2i+1In other words, the sub-pixels spatially corresponding thereto are not found in the display screen. Here, in order to make compatibility between two right and left adjacent pixels (R'_2iAnd R'_2i+2) By averaging the twoAnd sub-pixel R of image_2i+1A comparison is made. An expression of the red channel gray level squared error can thus be obtained.

Similarly, an expression of the square error of the blue channel gray scale can be obtained:

(II) color edge effect

The visual effect of the displayed image on the RGBG AMOLED screen will degrade due to the effect of the color side effect, and therefore this effect should be minimized as much as possible during the RGB to RGBG conversion. The color edge effect in the RGBG panel is shown in fig. 3, for a specific example, for the high frequency component in the original image (the gray scale difference between the adjacent pixels is large), for example, the gray scale of the original image is:

0(0,0,0),1(255,255,255),2(255,255,255),3(0,0,0)，

if the gray scale of RGBG is obtained by direct assignment

B_i-1'＝B_i-1＝0,G_i-1'＝G_i-1＝0

R_i'＝R_i＝255,G_i'＝G_i＝255

B_i+1'＝B_i+1＝255,G_i+1'＝G_i+1＝255

R_i+2'＝R_i+2＝0,G_i+2'＝G_i+2＝0。

The gray scale of the virtual pixel is: 0"(255,0,0),1"(255,255,255),2"(0,255,255). In the original image, the color of pixel 0 is black, and after conversion, the corresponding pixel becomes red, which causes a very large color distortion. For some high frequency image information, the converted color is different from the color of itself, and this phenomenon is called color fringing effect.

To evaluate the color side effect, the original image is first converted into YUV space, where U and V represent color-related information, a histogram of U and V is obtained, the virtual image is then converted into YUV space, a histogram is obtained, and then the difference between the original image and the histogram of the virtual image is compared, as an example of evaluation, the following equation is given:

PSNR_U＝10×log₁₀(255²/MSE_U)

wherein, Ui and Ui' respectively represent U components of the original image and the virtual image, and N is the total number of pixels of the image. PSNR is defined in a similar way_V。PSNR_UThe larger the difference between the original image and the virtual image is, the smaller the difference is, and thus, the better the display effect on RGBG is.

In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The first embodiment of the invention relates to a pixel structure conversion method of a display screen. Fig. 1 is a flow chart of the pixel structure conversion method.

Specifically, as shown in fig. 1, the pixel structure conversion method includes the steps of:

in step S101, three primary color gray scales of all pixel units in the input RGB format image are extracted.

Specifically, for an input image, three primary color gray scales of all pixel units in the image are extracted and stored.

And then, step S102 is carried out, three primary color gray scales of three to five adjacent pixel units in the original image are taken out one by one and are converted into new gray scales, and the gray scale square difference between the original pixel unit and the new pixel unit is enabled to be minimum.

Specifically, three primary color gray scales of adjacent three to five pixel units in the image are taken out one by one. The expressions of the new gray levels of the pixel units are set so that the gray level difference between the original pixel unit (original image in RGB format) and the new pixel unit (RGBG format) is minimized. Preferably, three primary color gray scales of adjacent four pixel units can be extracted, and the calculation amount is small and the display precision is high.

In the image to be displayed and the display screen, the green channel sub-pixels are in one-to-one correspondence, so the gray scale of the green channel is not changed by the algorithm. However, for the red and blue channels, the number of sub-pixels in the display screen is half the number of sub-pixels in the image. Therefore, a new sub-pixel gray scale cannot be obtained by direct assignment. For the gray scales of the red channel and the blue channel of the display screen, the gray scales of the red channel and the blue channel in the display screen can be set as unknowns, a function of gray scale square difference changing along with the unknowns is constructed, and then the corresponding gray scale when the gray scale square error is minimum is solved.

Order toWhere k is 0, 1., N/2-1, resulting in N/2-ary 1-order equations, which have unique solutions.

Here, the red channel is explained as an example. The gray level square error of the red channel is

Then, an expression of the partial derivative is found. Due to the fact that in the function F_ROf all the summation terms of (c), two will occur R'_2k

(1) When the value of i is equal to k,in the item, R 'appears'_2k

(2) When i is equal to k-1,in the item(s),

that is to sayR may also occur_2kAnd therefore, the first and second electrodes are,

wherein k is 0,1

Written in matrix form:

solving this matrix equation yields:

for the blue channel, the same reasoning can be used

After the new gray scale values are determined, the RGB to RGBG conversion can be performed.

Thereafter, the process proceeds to step S103, and after all the new gray scales are obtained, the input image is displayed again in the RGBG manner.

Specifically, for each pixel unit taken out, the gray scale R of the pixel unit at the corresponding position in the RGBG screen is obtained_2i'、G_2i' and B_2i+1'、G_2i+1' and storing, and then converting the converted new image gray scale R according to the image display driving time sequence of the RGBG type AMOLED screen_2i'、G_2i'、B_2i+1'、G_2i+1' to the screen side of the display,

this flow ends thereafter.

Effect

For objective evaluation of color-edge effects, PSNR is compared for different images_UAnd PSNR_VAs shown in tables 1 and 2.

As can be seen from tables 1 and 2, the PSNR value of the image processed by the invention is always larger than that of the image processed by the prior art, which shows that the invention greatly improves the color edge effect.

TABLE 1 PSNR of three different images_UComparison of

Image of a person	Prior Art	The invention
			(a)	32.79	32.93
(b)	29.30	30.67
			(c)	31.83	32.54

TABLE 2P of three different imagesSNR_VComparison of

Image of a person	Prior Art	The invention
			(a)	32.20	32.15
(b)	30.25	31.34
			(c)	32.26	32.80

To evaluate the difference between the original image and the processed result image, FB and FR need to be compared, taking the average of all pixel units, MFB and MFR respectively.

TABLE 3 MFB comparison of three images

Image of a person	Prior Art	The invention
			(a)	54.96	47.11
(b)	157.89	115.92

(c)

49.89

44.84

TABLE 4 MFR comparison of the three images

Image of a person	Prior Art	The invention
			(a)	58.41	51.49
(b)	141.52	103.84
			(c)	50.39	45.31

The MFB and MFR values for the same three images processed by the pixel conversion algorithm are shown in tables 3 and 4. The value of the image processed by the invention is always smaller than the processing result of the prior art. It can also be seen that the pixel conversion method of the present invention is superior to the prior art by processing a large number of other images and analyzing the results.

In the above description of the present invention, the AMOLED panel is taken as an example, but the present invention is applicable to all display panels with RGBG pixel structures, and is not limited to the AMOLED panel. That is, the above-described pixel structure conversion method of the present invention can be applied to all display panels of BGRG pixel structures, without being limited to AMOLED panels, as long as it is simply replaced.

The method embodiments of the present invention may be implemented in software, hardware, firmware, etc. Whether the present invention is implemented as software, hardware, or firmware, the instruction code may be stored in any type of computer-accessible memory (e.g., permanent or modifiable, volatile or non-volatile, solid or non-solid, fixed or removable media, etc.). Also, the Memory may be, for example, Programmable Array Logic (PAL), Random Access Memory (RAM), Programmable Read Only Memory (PROM), Read-Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), a magnetic disk, an optical disk, a Digital Versatile Disk (DVD), or the like.

The second embodiment of the invention relates to a pixel structure conversion device of a display screen. Fig. 5 is a schematic structural diagram of the pixel structure conversion device. The actual structure of the present invention can be adjusted as needed, and is not limited to the structure in fig. 5.

Specifically, the pixel structure conversion device performs pixel conversion, and solves the problem of mismatching of pixel structures. As shown in fig. 5, the pixel structure conversion apparatus 100 includes:

a gray scale obtaining section 101 for taking out three primary color gray scales of all pixel units in an input RGB-format image;

a gray scale conversion part 102, configured to take out three primary color gray scales of three to five adjacent pixel units in the original image one by one, and convert the three primary color gray scales into a new gray scale, so that a gray scale square difference between the original pixel unit and the new pixel unit is minimum; and

and a display unit 103 for displaying the input image again in the RGBG mode after all the new gradations are obtained.

The first embodiment is a method embodiment corresponding to the present embodiment, and the present embodiment can be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the first embodiment.

It should be noted that, each unit mentioned in each device embodiment of the present invention is a logical unit, and physically, one logical unit may be one physical unit, or may be a part of one physical unit, or may be implemented by a combination of multiple physical units, and the physical implementation manner of these logical units itself is not the most important, and the combination of the functions implemented by these logical units is the key to solve the technical problem provided by the present invention. Furthermore, the above-mentioned embodiments of the apparatus of the present invention do not introduce elements that are less relevant for solving the technical problems of the present invention in order to highlight the innovative part of the present invention, which does not indicate that there are no other elements in the above-mentioned embodiments of the apparatus.

It is to be noted that in the claims and the description of the present patent, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, the use of the verb "comprise a" to define an element does not exclude the presence of another, same element in a process, method, article, or apparatus that comprises the element.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims (6)

1. A pixel structure conversion method of a display screen is characterized by comprising the following steps:

taking out three primary color gray scales of all pixel units in the input RGB type image;

three primary color gray scales of three to five adjacent pixel units of the input image are taken out one by one and converted into new gray scales, so that the gray scale square error between the original pixel unit and the new pixel unit is minimum,

wherein the conversion to a new gray scale comprises the steps of: for the red channel and the blue channel, a function of which the gray scale square difference changes along with the gray scale of the red channel and the blue channel in the display screen is constructed, and the corresponding gray scale when the gray scale square difference is minimum is solved; meanwhile, for the green channel, the new gray level is equal to the gray level of the original image,

wherein, for the red channel, the gray scale squared error is:

wherein N is the total pixel number in the image, R'_2iAnd R'_2i+2Is the unknown to be solved, F_RIs with R'_2iAnd R'_2i+2Function of variation, R_2i、R_2i+1The gray scale of the original image;

for the blue channel, the expression of the gray level squared error is:

n is the total number of pixels in the image, B'_2i-1And B'_2i+1Is the unknown to be solved, F_BIs with B'_2i-1And B'_2i+1Function of variation, B_2iAnd B_2i+1The gray scale of the original image; and

after all new gray scales are obtained, the input image is displayed again in an RGBG mode; wherein,

the matrix form of the gray level squared error of the red channel is:

the matrix form of the gray level square error of the blue channel is as follows:

wherein,

2. the pixel structure conversion method according to claim 1,

the display screen is an AMOLED screen.

3. The pixel structure conversion method according to claim 1,

when three primary color gray scales of three to five adjacent pixel units of the input image are taken out one by one, the three primary color gray scales of four pixel units are taken out.

4. A pixel structure converting device for a display panel, comprising:

a gray scale obtaining part for taking out three primary color gray scales of all pixel units in the input RGB type image;

the gray scale conversion part is used for taking out three primary color gray scales of three to five adjacent pixel units of the input image one by one and converting the three primary color gray scales into new gray scales to ensure that the gray scale square difference between the original pixel unit and the new pixel unit is minimum, wherein the conversion into the new gray scales comprises the following steps: for the red channel and the blue channel, a function of which the gray scale square difference changes along with the gray scale of the red channel and the blue channel in the display screen is constructed, and the corresponding gray scale when the gray scale square difference is minimum is solved; meanwhile, for the green channel, the new gray level is equal to the gray level of the original image,

wherein, for the red channel, the gray scale squared error is:

wherein N is the total pixel number in the image, R'_2iAnd R'_2i+2Is the unknown to be solved, F_RIs with R'_2iAnd R'_2i+2Function of variation, R_2i、R_2i+1The gray scale of the original image;

for the blue channel, the expression of the gray level squared error is:

n is the total number of pixels in the image, B'_2i-1And B'_2i+1Is the unknown to be solved, F_BIs with B'_2i-1And B'_2i+1Function of variation, B_2iAnd B_2i+1Is the gray level of the original image, wherein,

the matrix form of the gray level squared error of the red channel is:

the matrix form of the gray level square error of the blue channel is as follows:

wherein,

and

and a display unit for displaying the input image again in the RGBG mode after all new gray scales are obtained.

5. The pixel structure converting device according to claim 4,

the display screen is an AMOLED screen.

6. The pixel structure converting device according to claim 4,

when three primary color gray scales of three to five adjacent pixel units of the input image are taken out one by one, the three primary color gray scales of four pixel units are taken out.