KR101566201B1 - Method of converting WRGB to RGB and device for executing the same - Google Patents

Abstract

The present invention provides a method for converting a WRGB color filter array into an RGB color filter array so as to be easily applicable to a general-purpose digital camera, and an image processing apparatus for executing the method. The WRGB-RGB conversion method of the present invention comprises the steps of (a) selecting a pixel group composed of a predetermined number of pixels from a WRGB array, and classifying the pixels of the selected pixel group into a plurality of reference blocks; (b) calculating a block weight of each of the reference blocks; (c) converting RGB included in each of the reference blocks into hue, saturation, and brightness; (d) creating a lookup table using the converted hue and saturation; And (e) calculating RGB color ratios of W pixels included in the selected pixel group by referring to the lookup table.

Description

TECHNICAL FIELD [0001] The present invention relates to a WRGB-RGB conversion method,

The present invention relates to an image processing method, and more particularly, to a method for converting a WRGB (White Red Green Blue) color filter array into an RGB color filter array and an image processing apparatus for executing the method.

2. Description of the Related Art In recent years, as the spread of mobile communication devices such as mobile phones has rapidly increased, the demand for cameras mounted on mobile communication devices is also rapidly increasing. The camera includes an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) to photograph a subject and convert the photographed image into a signal to perform image processing.

The image sensor has a plurality of pixels, and the plurality of pixels typically have a color array structure of a Red Green Blue (Bayer) pattern. The plurality of pixels convert an optical signal incident upon a subject into an electrical signal. The plurality of pixels detect values for one of R (Red), G (Green) and B (Blue), respectively. The R, G, and B values detected in each pixel are converted into digital signals by an ADC (Analog Digital Converter) provided in the image sensor and transmitted to the image signal processing device. The image signal processing apparatus reproduces the color of the object by combining R, G, and B included in the received signal.

At this time, since the signal detected by each pixel includes only information on one of R, G, and B, the image signal processing apparatus can detect the R, G, and B information And performs color interpolation processing so as to include both of them. That is, the image signal processing apparatus performs a process of interpolating a signal output from the image sensor into a signal capable of displaying an image in a display device such as a liquid crystal display (LCD) or a light-emitting diode (LED) display.

However, in recent years, image sensors having color arrangements different from RGB Bayer pattern, for example, a color arrangement of WRGB (White Red Green Blue) have been used. In the case of an image sensor using a WRGB array, it is not easy to apply to a general-purpose digital camera that receives and processes signals of an RGB Bayer pattern.

The present invention provides a method for converting a WRGB color filter array into an RGB color filter array so as to be easily applicable to a general-purpose digital camera, and an image processing apparatus for executing the method.

According to an aspect of the present invention,

A method of converting a WRGB array into an RGB array, comprising the steps of: (a) selecting a pixel group composed of a predetermined number of pixels from the WRGB array; and classifying the pixels of the selected pixel group into a plurality of reference blocks; (b) calculating a block weight of each of the reference blocks; (c) converting RGB included in each of the reference blocks into hue, saturation, and brightness; (d) creating a lookup table using the converted hue and saturation; And (e) calculating RGB color ratios of W pixels included in the selected pixel group by referring to the lookup table.

In order to solve the above problems,

A method for converting a WRGB array into an RGB array, comprising the steps of: (a) calculating color correlation coefficients having a color correlation between W pixels, R pixels, G pixels and B pixels among the pixels constituting the WRGB array; Creating an edge direction map of pixels; And (b) estimating a green value of the W pixel using the color correlation coefficients and the edge direction map.

In order to solve the above problems,

(a) selecting a pixel group composed of a predetermined number of pixels from the WRGB array and classifying the pixels of the selected pixel group into a plurality of reference blocks, (b) assigning a weight of each of the reference blocks, (C) converting RGB included in each of the reference blocks into hue, saturation, and brightness; (d) performing a lookup operation using the converted hue and saturation, And (e) calculating the RGB color ratios of the W pixels included in the selected pixel group by referring to the lookup table, and sequentially performing W-RGB conversion for converting the WRGB array into RGB arrays part; And a color interpolation unit interpolating RGB pixels output from the W-RGB conversion unit into pixels having red, green, and blue colors, respectively.

In order to solve the above problems,

(a) calculating color correlations between W pixels, R pixels, G pixels and B pixels among the pixels constituting the WRGB array and creating an edge direction map of the pixels; (b) Estimating a green value of a W pixel by using color correlation coefficients and an edge direction map, and sequentially converting the WRGB array into an RGB array; And a color interpolator for interpolating RGB pixels output from the W-G converter into pixels having red, green, and blue colors, respectively.

As described above, according to the present invention, the WRGB color filter array is converted into an RGB array, that is, an RGB Bayer pattern.

Therefore, it is easy to apply to a general-purpose digital camera using an RGB Bayer pattern.

1 is a flowchart of a WRGB-RGB conversion method according to an embodiment of the present invention.
Figure 2 shows a portion of a WRGB color filter array.
3 is a view for explaining a method of classifying a specific filter group of the WRGB color filter array into a plurality of reference blocks.
4 is a flowchart of a WRGB-RGB conversion method according to another embodiment of the present invention.
5 is a diagram showing four pixels of the WRGB color filter array.
6 is a view for explaining the edge direction of the WRGB color filter array.
Figure 7 shows the spectra reproduced using the present invention and the prior art.
8 is a schematic block diagram of an image processing apparatus according to an embodiment of the present invention.
9 is a schematic block diagram of an image processing apparatus according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. Like reference numerals in the drawings denote like elements.

1 is a flowchart of a WRGB-RGB conversion method according to an embodiment of the present invention. Referring to FIG. 1, the WRGB-RGB conversion method includes first through fifth steps. The WRGB-RGB conversion method according to an embodiment of the present invention is a method of converting W pixels included in a WRGB color filter array into R (Red), G (Green), and B (Blue).

In a first step 111, a pixel group composed of nine pixels around a W pixel is selected, and the selected pixel group is classified into a plurality of reference blocks. That is, as shown in FIG. 3, a pixel group composed of nine pixels arranged in a rectangle around a W (White) pixel is selected, and then a W pixel is subtracted from nine pixels included in the pixel group The eight pixels are classified into four reference blocks. In this case, the four reference blocks are composed of three pixels each including one red pixel, one green pixel, and one blue pixel.

As shown in FIG. 2, the WRGB color filter array includes a plurality of W pixels, a plurality of R pixels, a plurality of G pixels, and a plurality of B pixels, and the WRGB color filter array is regularly arranged at upper, have.

That is, the WRGB color filter array can be classified into the following four pixel groups according to the arrangement method of the pixels.

(1) A group of pixels whose R, G, and B pixels are arranged in a rectangle around a W pixel.

(2) A group of pixels whose W, G, and B pixels are composed of rectangles around an R pixel.

(3) A group of pixels composed of W, R, and B pixels around a G pixel.

(4) A group of pixels whose W, R, and G pixels are formed by squares around B pixels.

In the present invention, a pixel group of the (1) -th mode is selected and used among the four pixel groups.

The W pixel is a sum of all three colors (Red, Green, Blue), and these three colors can be defined as shown in the following Equation (1).

[Equation 1]

BW _{1 to} BW ₄ represent the weight of each of the four reference blocks classified as CR _{1 to} CR ₄ and CR _{4 to} CR ₄ , ~CR _{G1 G4,} CR ~CR _{B1 B4} represents a color ratio (ratio) with the each of the four reference blocks.

In a second step 121, a block weight of each of the reference blocks is calculated. The weight (BW _i ) can be calculated using the following equation (2).

&Quot; (2) "

Here, sum _i is a sum of values of R pixel, G pixel and B pixel included in the corresponding reference block, white represents a value of W pixel, and sum _j represents a sum of R Represents the sum of the values of the pixels, the G pixels, and the B pixels. That is, if there are four reference blocks, the weight is calculated for each reference block, so a total of four weights are required.

In the third step 131, RGB of the reference blocks are converted into H (Hue), S (Saturation), and V (Value). The following Equations 3 and 4 can be used to convert RGB to HSV.

Generally, when changing the brightness of a pixel, the color information of the pixel does not change. That is, the color information of a pixel is not affected by brightness and is affected by hue and saturation. Therefore, in the present invention, the hue is not considered, and the hue and saturation including the color information of the pixel are calculated.

The color h may be calculated using Equation 3 below, and the saturation s may be calculated using Equation 4 below.

&Quot; (3) "

&Quot; (4) "

In a fourth step 141, a lookup table is created. That is, a lookup table is created using the hue and saturation calculated in the third step. The index for creating the look-up table is configured as shown in Table 1 below.

beat Meaning of code 1-2 01: R is the maximum, 10: G is the maximum, 11: B is the maximum,
00: RGB is all the same 3 1 if the other is greater than the first, or 0 4 If RGB is all zeros, 1, or 1 5 to 10 The value of H (color) (decimal) 11-16 The value of S (saturation) (decimal number)

An example of the index created using Table 1 is as follows.

1011100001110000

Here, the value "10 " constituting one or two bits (the first and second bits on the left) indicates that G is the maximum,

A value "1" constituting 3 bits (third bit from the left) indicates that R is greater than B,

A value "1" constituting 4 bits (fourth bit from the left) indicates that S is not zero,

A value "100001" constituting 5 to 10 bits (fifth to tenth bits from the left) indicates that the value of H is "65 &

The value "110000 " constituting 11 to 16 bits (the eleventh and sixteenth bits from the left) indicates that the value of S is" 96 ".

In a fifth step 151, RGB color ratios of W pixels are calculated. The RGB color ratios (FCR _R , FCR _G , and FCR _B ) of the W pixel can be calculated using Equation (5) below.

&Quot; (5) "

CR _Ri is the color ratio of the R pixels included in the corresponding reference block, CR _Gi is the color ratio of the G pixels included in the corresponding reference block, CR _Bi is the color ratio of the B pixels included in the corresponding reference block, And BW _i represents the weight of the corresponding reference block.

CR _Ri , CR _Gi and CR _Bi can be calculated by the following equation (6).

&Quot; (6) "

Here, Wi denotes a value of W pixel, R _i denotes a value of R pixel of the corresponding reference block, G _i denotes a value of G pixel of the corresponding reference block, B _i denotes a value of B pixel of the corresponding reference block Respectively.

The values (Whitex) of all the W pixels included in the WRGB color filter array can be calculated using Equation (7) using the color ratios calculated by Equation (5).

&Quot; (7) "

Whitex = White Signal x FCRx

Using the values calculated by Equation (7), the W pixel can be converted into a pixel having red, green, and blue.

As described above, according to an embodiment of the present invention, the WRGB color filter array can be converted into an RGB color filter array (Bayer pattern). Accordingly, it is easy to use an image sensor using a WRGB color filter array in a general purpose camera using an RGB Bayer pattern. This improves the accuracy of image reproduction.

4 is a flowchart of a WRGB-RGB conversion method according to another embodiment of the present invention. Referring to FIG. 4, the WRGB-RGB conversion method includes first and second steps. A color conversion method according to another embodiment of the present invention is a method of converting a W pixel included in a WRGB color filter array into a G (Green) pixel.

In a first step 411, a color correlation between the W pixel, the R pixel, the G pixel, and the B pixel is calculated and an edge direction map of the pixels is created.

First, a method of calculating a color correlation between a W pixel, an R pixel, a G pixel, and a B pixel will be described.

The value W of the W pixel can be defined by the following equation (8).

&Quot; (8) "

W =? R +? G +? B

Here, R represents the value of R pixel, G represents the value of G pixel, and B represents the value of B pixel. Further,?,?, And? Are correlation coefficients indicating the correlation between the R pixel, the G pixel, and the B pixel, respectively.

Equation 8 is derived using a 2x2 matrix structure consisting of one W pixel, one R pixel, one G pixel, and one B pixel, as shown in Fig.

The correlation coefficients alpha, beta, and gamma can be accurately calculated using Least Square of the non-singular situation shown in Equation 9 below.

&Quot; (9) "

Equation (9) can be expressed by Equation (10) below.

&Quot; (10) "

Ax = b

Here, A, x and b can be expressed by the following equation (11).

&Quot; (11) "

The values of the correlation coefficients (?,?,?) Can be obtained using Equations (10) and (11).

Next, a method of creating an edge direction map will be described. To create an edge direction map, first create an edge strength map. An example of an edge intensity map is shown in Table 2 below.

S _1,1 S _1,2 S _1,3 S _1,4 S _1,5 S _2.1 S _2.2 S _2,3 S _2,4 S _2,5 S _3,1 S _3,2 S _3,3 S _3,4 S _3,5 S _4,1 S _4,2 S _4,3 S _4,4 S _4,5 S _5,1 S _5,2 S _5,3 S _5,4 S _5.5

In the edge intensity map, S _{i, j} represents the edge strength of the corresponding cell. The edge intensity of each pixel can be calculated using the following equation (12).

&Quot; (12) "

S _{i , j} =

+

The formula for calculating one of the edge intensities (S _{2 , 2} ) (corresponding to the edge intensity of W pixels) of the edge intensities of the pixels included in the edge intensity map using Equation (12) is shown in Equation (13).

&Quot; (13) "

S _{2 .2} =

+

The edge directions of each pixel can be confirmed using the edge intensities calculated using Equation (13). Each pixel has four edge directions (45 degrees, 135 degrees, 225 degrees, 315 degrees), as shown in Fig. At this time, an edge at 45 degrees value (D_45 _{i, j)} and the 135-degree edge values (D_135 _{i, j)} in the direction of orientation can be found by using the equation (14) below.

&Quot; (14) "

D_45 _{i, j} =

D-135 _{i, j} =

If the edge value (D_45 _{i, j} ) in the 45-degree direction is larger than the edge value (D_135 _{i, j} ) in the 135-degree direction as a result of the calculation of Equation (14) _{, the} corresponding pixel has an edge value in the 45- If the edge value (D_135 _{i, j} ) in the direction of the arrow is greater than the edge value (D_45 _{i, j} ) in the direction of 45 degrees _{, the} corresponding pixel has an edge value in the 135 degree direction. An example of an edge direction map in which an edge value is set is shown in Table 3 below.

45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 45 ° 135 ° 45 ° 45 ° 45 135 ° 135 ° 45 ° 45 ° 135 ° 135 ° 135 °

The edge direction of the W pixel can be confirmed using the edge direction map as shown in Table 3 above. By determining the edge direction, it is possible to determine which pixel to refer to when estimating the green value of the W pixel. That is, the green value of the W pixel is estimated by referring to the G pixels existing in the direction along the edge direction. By using the above method, blur is significantly reduced at the edge of the image than in the conventional method.

If the green value of the W pixel is estimated without using the edge direction map, a lot of blur occurs at the edge of the image. For example, when the edge direction of the W pixel is 135 degrees and the pixels in the 45 degree direction are referred to, the green value of the W pixel is an average value of the colors (e.g., red and yellow) of the pixels existing in the 45 degree direction, And the average value of yellow color). Thus, the blur increases significantly at the edges of the image.

In the present invention, the edge direction is only 45 degrees and 135 degrees. For that reason, the present invention converts a W pixel into a G pixel having a green value, and estimates a green value by referring to only G pixels existing around the W pixel. The G pixels around the W pixel exist only in the 45 degree direction and the 135 degree direction around the W pixel, as shown in FIG.

In the above-described first step, two operations (an operation of calculating a color correlation between a W pixel, an R pixel, a G pixel, and a B pixel, an operation of creating an edge direction map of the pixels) May proceed at the same time or sequentially, and the result is the same.

In a second step 421, the green value of the W pixel is estimated using the color correlation coefficients and the edge direction map. In order to estimate the green value of the W pixel, it is preferable to first estimate a rough green value, and finally estimate a precise green value using the estimated green value.

First, a method of estimating the approximate green value of the W pixel will be described.

The approximate green value G of the W pixel can be calculated using the following equation (15). Equation (15) corresponds to the case where the W pixel has an edge value in the 135 degree direction.

&Quot; (15) "

+ EG = A + E = W x

+ E

Where A is the mean value of two adjacent G pixels in the 135 degree direction, E is a non-linear error, W is the value of the W pixel,?,? And? Are the R pixel, And B pixel color correlation coefficients, R _average is the average value of R pixels around W pixel, G _average is _average value of G pixels around W pixel, B _average is _average value of B pixels around W pixel, G _{average And} 135 represents the average value of G pixels in the 135-degree direction around the W pixel.

The nonlinear error (E) can be calculated by the following equation (16).

&Quot; (16) "

Here, k represents an edge strength coefficient.

If the W pixel has an edge value in the direction of 45 degrees, the internal variables for the formula for obtaining the approximate green value of the W pixel differ from those described above.

From Equation (15) and Equation (16), the following Equation (17) can be derived. That is, the approximate green value (G_45 _{i, j} , G_135 _{i, j} ) of the W pixel can be calculated using the following Equation (17).

&Quot; (17) "

Here, G_45 _{i, j} is the green value when the W pixel has the edge value in the 45 degree direction, and G_135 _{i, j} is the green value when the W pixel has the edge value in the 135 degree direction. Accordingly, one of the two green values (G_45 _{i, j} , G_135 _{i, j} ) can be selected and used according to the edge direction map.

Next, a precise green value of the W pixel is estimated. The precise green value of the W pixel can be calculated using the following equation (18).

&Quot; (18) "

Where M _x is the direction weight and D _x is the direction strength. The direction weight (M _x ) and the direction strength (D _x ) can be calculated using the following equations (19) and (20).

&Quot; (19) "

M ₁ = D ₂ D ₃ D ₄

M ₂ = D ₁ D ₃ D ₄

M ₃ = D ₁ D ₂ D ₄

M ₄ = D ₁ D ₂ D ₃

&Quot; (20) "

Here, S _{i , j} is an edge strength that can be calculated using Equation (12).

As described above, by estimating the precise green value of the W pixel, the W pixel can be converted to a G pixel having a green value.

Thus, by estimating the green value of the W pixel, the WRGB color filter array can be converted to an RGB color filter array, that is, an RGB payload pattern. Therefore, it is easy to use an image sensor using a WRGB color filter array in a general-purpose camera using an RGB Bayer pattern. Further, since the present invention uses the edge direction to estimate the green value, the accuracy is improved near the edge, and the blur is significantly reduced at the edges of the image.

Figure 7 shows the spectra reproduced using the present invention and the prior art. In Fig. 7, the horizontal axis indicates the wavelength and the vertical axis indicates the value of the spectrum.

Referring to FIG. 7, it can be seen that the spectrum 721 reproduced by the present invention is more similar to the spectrum 711 than the spectrum 731 reproduced by the prior art.

As described above, when the image is reproduced using the WRGB-RGB conversion method according to the present invention, the accuracy of the image is remarkably improved as compared with the prior art.

8 is a schematic block diagram of an image processing apparatus according to an embodiment of the present invention. Referring to FIG. 8, the image processing apparatus 801 includes a W-RGB conversion unit 811 and a color interpolation unit 821.

The W-RGB conversion unit 811 performs a WRGB-RGB conversion method according to an embodiment of the present invention to convert W pixels included in the WRGB color filter array into R (Red), G (Green), and B Conversion. That is, the W-RGB conversion unit 811 selects (a) a predetermined number of pixel groups in the WRGB color filter array and classifies the selected pixel groups into a plurality of reference blocks, (b) (C) converting RGB included in each of the reference blocks into hue, saturation, and brightness; (d) (E) calculating the RGB color ratios of the W pixels included in the pixel group by referring to the lookup table, and sequentially calculating the WRGB color filter array as RGB Into a color filter array.

The detailed implementation method of each of the above steps is explained in detail through the WRGB-RGB conversion method according to the embodiment of the present invention shown in FIG. 1, and thus a duplicated description will be omitted.

The color interpolation unit 821 interpolates R pixels, G pixels, and B pixels output from the W-RGB conversion unit 811 into pixels including red, green, and blue, respectively.

Thus, the image processing apparatus 801 receives and processes the WRGB signal from a WRGB image sensor (not shown), and sends the processed signal to a display (not shown) such as an LCD or an LED to reproduce the image.

9 is a schematic block diagram of an image processing apparatus according to another embodiment of the present invention. Referring to FIG. 9, the image processing apparatus 901 includes a W-G conversion unit 911 and a color interpolation unit 921.

The W-G conversion unit 911 performs a WRGB-RGB conversion method according to another embodiment of the present invention to convert W pixels included in the WRGB color filter array into G (Green) pixels. That is, the WG converting unit 911 calculates (a) the color correlation between the W pixel, the R pixel, the G pixel and the B pixel of the WRGB color filter array and generates an edge direction map of the pixels And (b) estimating the green value of the W pixel by using the color correlation coefficients and the edge direction map, and sequentially converting the WRGB color filter array into the RGB color filter array.

The detailed implementation method of each of the above steps is described in detail through the WRGB-RGB conversion method according to another embodiment of the present invention shown in FIG. 4, and thus a duplicated description will be omitted.

The color interpolation unit 921 interpolates R pixels, G pixels, and B pixels output from the W-G conversion unit 911 into pixels having red, green, and blue colors, respectively.

Thus, the image processing apparatus 901 receives and processes the WRGB signal from the WRGB image sensor, and sends the processed signal to a display such as an LCD or an LED to reproduce the image.

Although the present invention has been described with reference to the embodiments shown in the drawings, it is to be understood that various modifications and equivalent embodiments may be made by those skilled in the art without departing from the scope of the present invention. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

Claims (18)

A method for converting a WRGB array to an RGB array,
(a) selecting a pixel group composed of a predetermined number of pixels from the WRGB array, and classifying the pixels of the selected pixel group into a plurality of reference blocks;
(b) calculating a block weight of each of the reference blocks;
(c) converting RGB included in each of the reference blocks into hue, saturation, and brightness;
(d) creating a lookup table using the converted hue and saturation; And
(e) calculating RGB color ratios of W pixels included in the selected pixel group by referring to the lookup table. The method according to claim 1,
And converting the W pixel into a pixel having red, green, and blue using the calculated RGB color ratios of the W pixel. The method according to claim 1,
Wherein the pixels included in the selected pixel group are composed of nine pixels around a W pixel. The method according to claim 1,
Wherein the reference blocks are classified into four, and each of the reference blocks includes one R pixel, one G pixel, and one B pixel. The method according to claim 1,
The weight is expressed by the following equation (

) Where sum _i is a sum of values of R pixels, G pixels and B pixels included in the corresponding reference block, white represents a value of W pixel, sum _j is included in the plurality of reference blocks Values of R pixels, G pixels, and B pixels, which are obtained by adding the values of R pixels, B pixels, and B pixels). The method according to claim 1,
The RGB color ratios of the W pixel may be expressed as: < RTI ID = 0.0 >

,

,

) Where CR _Ri is the color ratio of the R pixel included in the corresponding reference block, CR _Gi is the color ratio of the G pixel included in the corresponding reference block, CR _Bi is the color ratio of the B pixel included in the corresponding reference block And BW _i represents a weight of the corresponding reference block). A method for converting a WRGB array to an RGB array,
(a) calculating color correlation coefficients having a color correlation between W pixels, R pixels, G pixels and B pixels among the pixels constituting the WRGB array, and creating an edge direction map of the pixels; And
(b) estimating a green value of the W pixel using the color correlation coefficients and the edge direction map, and estimating a precise green value of the W pixel using the approximate green value The WRGB-RGB conversion method comprising the steps of: 8. The method of claim 7,
Estimating a green value of the W pixel, and then converting the W pixel into a G pixel having a green value using the estimated green value. 8. The method of claim 7,
Creating an edge strength map of the pixels, and creating the edge direction map using the edge strength map. 10. The method of claim 9, wherein the edge strength (S _{i, j} )
The equation (S _{i , j} =

+

) Where G _{i , j} are i and j th G pixels, R _{i , j} are i and j th R pixels, B _{i , j} are i and j th B pixels, And the WRGB-RGB conversion method. delete 8. The method of claim 7, wherein the approximate green value (G)
Equation (G = W x

+ E) (where, W denotes the value of the W pixel, E is a non-linear error, α, β and γ are each R pixels, G pixels and B deulyigo color correlation coefficient of the pixel, R _average is the surrounding W pixels R G _average is the average value of the G pixels around the W pixel, B _average is the average value of the B pixels around the W pixel, and G _{average.135 degrees} is the average value of the G pixels in the 135 degree direction around the W pixel In the WRGB-RGB conversion method. 13. The method of claim 12, wherein the nonlinear error (E)
Equation (

) (W (1), W (3), and W (5) represent W pixels arranged sequentially from the upper left to the lower right in the 135 degree direction) (k represents the edge intensity coefficient) WRGB-RGB conversion method characterized. 8. The method of claim 7,
Equation (

) And the equation (

), Where k is the edge intensity coefficient, M _x is the directional weight, W _{i, j} is the W pixel value, and α, β and γ are the color correlation coefficients of R pixel, G pixel, Where R _average is the mean value of R pixels around the W pixel, G _average.45 is the mean value of the G pixels in the 45 degree direction around the W pixel and B _average is the mean value of the B pixels around the W pixel) And the WRGB-RGB conversion method. 15. The method of claim 14, wherein the direction weight (M _x )
Formula _{(M 1 = D 2 × D} 3 × D 4, M 2 = D 1 × D 3 × D 4, M 3 = D 1 × D 2 × D 4, M 4 = D 1 × D 2 × D 3 ) (Where D _x represents the direction strength). 16. The method of claim 15, wherein the direction strength (D _x )
Equation (

,

,

,

) (Where S _{i, j} represents the edge strength). (a) selecting a pixel group composed of a predetermined number of pixels from the WRGB array and classifying the pixels of the selected pixel group into a plurality of reference blocks, (b) assigning a weight of each of the reference blocks, (C) converting RGB included in each of the reference blocks into hue, saturation, and brightness; (d) performing a lookup operation using the converted hue and saturation, And (e) calculating the RGB color ratios of the W pixels included in the selected pixel group by referring to the lookup table, and sequentially performing W-RGB conversion for converting the WRGB array into RGB arrays part; And
And a color interpolation unit that interpolates the RGB pixels output from the W-RGB conversion unit into pixels having red, green, and blue colors, respectively. (a) calculating color correlations between W pixels, R pixels, G pixels and B pixels among the pixels constituting the WRGB array and creating an edge direction map of the pixels; (b) Estimating the green value of the W pixel first by using the color correlations and the edge direction map and then estimating the precise green value of the W pixel by using the approximate green value, A WG conversion unit for converting the array into an RGB array; And
And a color interpolation unit interpolating the RGB pixels output from the WG conversion unit with pixels having red, green, and blue colors, respectively.

Images