KR101545064B1 - Apparatus for estimating pixel value of target pixel and method thereof - Google Patents

Abstract

The present invention relates to an apparatus and method for estimating a pixel value of a target pixel. A method of estimating a pixel value of a target pixel according to the present invention is a method of estimating a pixel value of a target pixel among N pairs of symmetric pixels arranged at the same angle around the center pixel, Selecting a symmetric pixel pair having a minimum sum of difference values from a brightness value Y and matching an angular value between a virtual straight line connecting the selected symmetric pixel pair and the reference axis to the center pixel; Selecting a center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel located between a plurality of center pixels having the angular values matched with each other, Calculating an average; (Y) of two intersecting pixels in which an imaginary straight line corresponding to the average angular value is extended about the target pixel and the extended straight line intersects with a virtual straight line between adjacent adjacent reference pixels, Obtaining a Cr, Cb value; And setting the brightness values (Y), Cr, and Cb values of the target pixel by averaging the brightness values (Y), Cr, and Cb values of the two intersecting pixels.
As described above, according to the present invention, by generating the interpolation pixel closest to the actual edge pixel in the vicinity of the edge of the image, it is possible to eliminate the noise due to the image enlargement and prevent the performance deterioration. In addition, the jigging phenomenon existing in the existing technology can be alleviated.

Description

[0001] APPARATUS FOR ESTIMATING PIXEL VALUE OF TARGET PIXEL AND METHOD THEREOF [0002]

The present invention relates to an apparatus and method for estimating a pixel value of a target pixel, and more particularly, to an apparatus and method for estimating a pixel value of a target pixel, which is one of interpolation pixels generated according to an image enlargement.

An image enlargement technique for enlarging a small image is widely used in various fields, and an interpolation scheme is widely used for such image enlargement. The enlarged image enlarged using the interpolation technique consists of interpolated pixels using reference pixels and reference pixels, which are pixels existing in the input image. Therefore, the closer the value of the interpolation pixels of the enlarged image is to the pixel value of the actual image, the better the quality of the enlarged image is.

Zero order interpolation, which is the simplest interpolation technique, has a disadvantage in that the image quality of the enlarged image is degraded. Spline interpolation, linear interpolation and cubic convolution interpolation have been proposed to improve the quality of the enlarged image, but there is a disadvantage that blurring occurs at the boundary of the image.

In particular, according to an image enlargement method for alleviating the jagging effect of a diagonal edge proposed in the A Novel Image Interpolation Method Using Bilateral Filter, a pixel value difference of a pair of pixels facing each other in a reference pixel around the position of an interpolation pixel to be generated , And the pixel value is set by finding the direction of the pixel value difference in the longest direction and calculating the average of pixel values in the corresponding direction. However, when such a method is used, noise is generated in the vicinity of the edge of the pixel thickness and the performance deterioration occurs severely.

The technique which is the background of the present invention is described in Korean Patent Laid-Open Publication No. 2008-0048385 (2008.06.02).

SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus and method for estimating a pixel value of a target pixel, which is one of interpolation pixels, with improved noise and performance degradation according to existing techniques, .

According to an aspect of the present invention, there is provided a method of estimating a pixel value of a target pixel located between a plurality of reference pixels having pixel values set therein, A sum of difference values of brightness values (Y) of the central pixel among the N pairs of symmetric pixels arranged at the same angle around the center pixel, which is one of the plurality of reference pixels, Selecting a minimum symmetric pixel pair and matching an angular value between a virtual straight line connecting the selected symmetric pixel pair and the reference axis to the center pixel; Selecting a center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel located between a plurality of center pixels having the angular values matched with each other, Calculating an average; (Y) of two intersecting pixels in which an imaginary straight line corresponding to the average angular value is extended about the target pixel and the extended straight line intersects with a virtual straight line between adjacent adjacent reference pixels, Obtaining a Cr, Cb value; And setting brightness values (Y), Cr, and Cb values of the target pixel by averaging the brightness values (Y), Cr, and Cb values of the two intersecting pixels.

In addition, the plurality of reference pixels may be spaced at equal intervals.

The matching of the angle value with the selected symmetric pixel pair to the center pixel may include covering a virtual window of a rectangular shape such that the reference pixel is positioned at each corner centering on the center pixel; Generating N virtual straight lines passing through the center pixel and evenly arranged at the same angle, selecting an intersecting pixel located at an intersection of the imaginary straight line and the virtual window; Linearly interpolating a brightness value (Y) of the crossing pixel using a brightness value (Y) of a surrounding reference pixel when the crossing pixel is not the reference pixel; Calculating a difference value between a first crossing pixel included in the N symmetric pixel pairs and a brightness value (Y) of the center pixel and a difference value between the remaining second crossing pixel and a brightness value (Y) of the center pixel, ; And selecting a minimum value among the N values generated by summing and matching an angle value between a virtual straight line connecting a pair of symmetric pixels corresponding to the selected minimum value and the reference axis to the center pixel.

The step of calculating the average of the angular values of the selected center pixel pair may include calculating a difference between the center value of the center pixel pair and the center value of the center pixel pair, The center pixel pair can be selected to calculate the average of the angular values of the selected center pixel pair.

The step of acquiring the brightness values (Y), Cr, and Cb values of the two intersecting crossing pixels may be performed such that when two intersecting pixels intersecting the imaginary straight line are not the reference pixels, (Y), Cr, and Cb values of the intersecting pixels are linearly interpolated using the Y, Cr, and Cb values, and when the two intersecting pixels are the reference pixels, the brightness value (Y) , Cr, and Cb values can be selected as the brightness values (Y), Cr, and Cb values of the two intersecting pixels.

The method may further include converting the brightness values (Y), Cr, and Cb values of the target pixel to R, G, and B values.

An apparatus for estimating a pixel value of a target pixel according to an embodiment of the present invention is an apparatus for estimating a pixel value of a target pixel positioned between a plurality of reference pixels having pixel values set therein, A symmetric pixel pair having a minimum sum of differences from the brightness value (Y) of the center pixel among N symmetric pixel pairs arranged at the same angle around the center pixel, A center pixel matching unit for matching an angular value between a virtual straight line connecting the selected symmetric pixel pair and a reference axis to the center pixel; Selecting a center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel located between a plurality of center pixels having the angular values matched with each other, An operation unit for calculating an average; (Y) of two intersecting pixels in which an imaginary straight line corresponding to the average angular value is extended about the target pixel and the extended straight line intersects with a virtual straight line between adjacent adjacent reference pixels, Cr, and Cb; And a setting unit for setting the brightness values (Y), Cr, and Cb values of the target pixel by averaging the brightness values (Y), Cr, and Cb values of the two intersecting pixels.

According to an apparatus and method for estimating a pixel value of a target pixel of the present invention, noise caused by enlargement of an image can be removed and performance deterioration can be prevented by generating an interpolation pixel closest to an actual edge pixel in the vicinity of an edge of the image. Also, the jagging phenomenon can be alleviated compared with the existing technology.

1 is a block diagram showing a configuration of an apparatus for estimating a pixel value of a target pixel according to an embodiment of the present invention.
2 is a block diagram illustrating a configuration of a center pixel matching unit according to an embodiment of the present invention.
3 is a flowchart of a pixel value estimation method of a target pixel according to an embodiment of the present invention.
4 is a detailed flowchart of step S310 of FIG.
5 is a diagram for explaining a process of obtaining an angle value of a center pixel in step S310 of FIG.
FIG. 6 is a diagram for explaining a linear interpolation process in step S314 of FIG. 3. FIG.
7 is a view for explaining the step S320 of FIG.
8 is a view for explaining the step S330 of FIG.
9 is an exemplary view showing an original image before image enlargement.
10 is an enlarged view of an original image of FIG. 9 enlarged by an existing method.
11 is a view showing an image obtained by enlarging an original image of FIG. 9 by a method according to an 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. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

First, an apparatus for estimating a pixel value of a target pixel according to an embodiment of the present invention will be described.

1 is a block diagram showing a configuration of an apparatus for estimating a pixel value of a target pixel according to an embodiment of the present invention.

1, an apparatus 100 for estimating a target pixel pixel value according to an exemplary embodiment of the present invention includes a central pixel matching unit 110, an operation unit 120, a control unit 130, and a setting unit 140 .

Herein, the target pixel means an interpolation pixel which is compared with a reference pixel (pixel value set pixel) which is an existing pixel constituting an image, and in more detail, a target pixel to be interpolated ). Particularly, in the embodiment of the present invention, the interpolation pixel means a pixel which should be interpolated before or after interpolation or the interpolation process is completed.

In addition, the apparatus 100 for estimating a pixel value of a target pixel according to an embodiment of the present invention can estimate the unknown pixel value of a target pixel, which is a target pixel to be interpolated, and thus can be implemented as an image enlargement apparatus.

First, the center pixel matching unit 110 selects one of N symmetric pixel pairs arranged at the same angle around the center pixel as a center pixel, which is any one of a plurality of reference pixels spaced apart at regular intervals, And selects the symmetric pixel pair whose sum of difference values with the brightness value (Y) is the minimum. Then, an angle value between a virtual straight line connecting the selected symmetric pixel pair and the reference axis is matched to the center pixel.

The arithmetic unit 120 selects a center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel located between a plurality of center pixels matched with angular values, And calculates the average of the angular values.

The control unit 130 extends a virtual straight line corresponding to the average angular value about the target pixel and calculates a brightness value of two intersecting pixels in which the extended straight line intersects with a virtual straight line between neighboring adjacent reference pixels (Y), Cr, and Cb.

The setting unit 140 sets the brightness values Y, Cr, and Cb of the target pixel by averaging the brightness values Y, Cr, and Cb of the two intersecting pixels.

In addition, the conversion unit (not shown) may convert the brightness values (Y), Cr, and Cb values of the target pixel into R, G, and B values.

Here, the brightness value (Y) means luminance information, and Cr and Cb mean chrominance-red and chrominance-blue, which are chrominance information, respectively.

2 is a block diagram illustrating a configuration of a center pixel matching unit according to an embodiment of the present invention.

The center pixel matching unit 110 may include a virtual processing unit 111, an intersection pixel selection unit 112, a linear interpolation unit 113, a summation unit 114, and an angle value matching unit 115.

The virtual processing unit 111 puts a virtual window of a rectangular shape around the center pixel so that the reference pixel is positioned at each corner.

The crossover pixel selection unit 112 generates N imaginary straight lines passing through the center pixel and evenly arranged at the same angle, and selects an intersecting pixel located at the intersection of the imaginary straight line and the imaginary window.

The linear interpolator 113 linearly interpolates the brightness value Y of the crossing pixel using the brightness value Y of the surrounding reference pixel when the crossing pixel is not the reference pixel.

The adder 114 adds the difference value between the first crossing pixel included in the N symmetric pixel pairs and the brightness value Y of the center pixel and the difference value between the remaining second crossing pixel and the brightness value Y of the center pixel Respectively.

The angular value matching unit 115 selects a minimum value among the N values generated by the summing, and matches an angle value between a virtual straight line connecting a pair of symmetric pixels corresponding to the selected minimum value and the reference axis to the center pixel.

Hereinafter, a method of estimating a pixel value of a target pixel according to an embodiment of the present invention will be described in detail with reference to FIG. 3 to FIG.

3 is a flowchart of a pixel value estimation method of a target pixel according to an embodiment of the present invention.

First, the center pixel matching unit 110 calculates a brightness value (Y) of the center pixel among N symmetric pixel pairs arranged at the same angle around the center pixel, which is one of the plurality of reference pixels, And the angular value between a virtual straight line connecting the selected pair of symmetric pixels and the reference axis is matched with the center pixel at step S310. Here, the plurality of basic pixels are spaced at equal intervals.

Hereinafter, the step S310 will be described in more detail with reference to FIG. 4 and FIG.

FIG. 4 is a detailed flowchart of step S310 of FIG. 3, and FIG. 5 is a diagram for explaining a process of obtaining an angle value of a center pixel in step S310 of FIG.

First, the virtual processor 111 puts a virtual window of a rectangular shape around the center pixel P0 so that the reference pixel is positioned at each corner (S311).

In FIG. 5, the reference pixel means a pixel having a pixel value set as a pixel existing in the original image, and the interpolation pixel means a pixel to be interpolated that occurs between the reference pixels due to enlargement of an original image or the like. The center pixel P0 is a pixel included in the reference pixel and is referred to as a center pixel for the sake of explanation for the reference pixel to which the angle value (edge angle value) is to be measured.

As shown in Fig. 5, the virtual window is formed in a rectangular shape, and the reference pixels located at each corner are A4, A12, B4 and B12. That is, the virtual window is formed in a rectangular shape having the reference pixel as an edge.

The cross pixel selection unit 112 selects 16 virtual straight lines A0-B0, A1-B1, A2-B2, ..., A14-B14, A15-B15 that pass through the center pixel P0 and are evenly arranged at the same angle And selects an intersecting pixel located at the intersection of the sixteen virtual straight lines and the imaginary window (S312). In FIG. 5, sixteen virtual straight lines are exemplified as one embodiment of the present invention, but the number of virtual straight lines can be changed according to the positions of pixel intervals or reference pixels. That is, a virtual straight line starting from the pixel B0 to the pixel B15 passes through the center pixel P0 in common, and intersects with the pixel A0 to the pixel A15 again. The pixels intersecting the sixteen virtual straight lines include a plurality of reference pixels B0, B4, B8, B12, B15, A4, A8 and A12 on the virtual window in addition to the center pixel P0 and interpolation pixels B1 to B3 , B5 to B7, B9 to B11, B13 to B15, A1 to A3, A5 to A7, A9 to A11 and A13 A15. That is, the intersecting pixels correspond to a plurality of reference pixels and interpolation pixels, and in FIG. 5, 32 intersecting pixels are generated.

The linear interpolator 113 determines whether the intersecting pixel corresponds to the reference pixel (S313).

If the intersection pixel is not the reference pixel, the linear interpolator 113 linearly interpolates the brightness value Y of the intersection pixel using the brightness value Y of the surrounding reference pixel (S314). That is, the linear interpolator 113 generates a brightness value Y of a pixel, which is an intersecting pixel and an interpolation pixel, by using a linear interpolation method with a brightness value Y of a surrounding reference pixel.

FIG. 6 is a diagram for explaining a linear interpolation process in step S314 of FIG. 3. FIG.

6, when the brightness value Y of the surrounding reference pixel A0 is 0.1 and the brightness value Y of the adjacent reference pixel A4 is 0.18, the brightness of the interpolation pixel A1, A2, and A3 The brightness value (Y) can be interpolated by linear interpolation to values of 0.12, 0.14, and 0.16, respectively.

The summing unit 114 calculates a difference value between the brightness value Y of the first crossing pixel and the center pixel included in the 16 symmetric pixel pairs and a difference value between the remaining second crossing pixel and the brightness value Y of the center pixel (S315).

For convenience, A0 to A15 cross pixels are referred to as first cross pixels, and B0 to B15 cross pixels, which are symmetrically paired with A0 to A15 cross pixels, are referred to as second cross pixels.

That is, the difference between A0 and the brightness value (Y) of the center pixel and the difference between B0 and the brightness value (Y) of the center pixel are summed up, and the difference between A1 and the brightness value (Y) And adds the difference in brightness value (Y). Thus, the difference between the brightness values (Y) of the 16 first cross pixels and the center pixel and the difference between the second cross pixels and the brightness value (Y) of the center pixel are summed to obtain 16 candidate values in total.

Here, the difference value of the brightness (Y) between the pixels is an absolute value form.

The angular value matching unit 115 selects a pair of symmetric pixels having the smallest value among the 16 candidate values generated by the summing and matches an angle value between the virtual straight line connecting the selected pair of symmetric pixels with the reference axis to the center pixel (S316).

For example, assuming that the pair of A6 and B6 pixels among the 16 pairs of symmetric pixels in Fig. 5 correspond to the minimum value, the angle value matching unit 115 determines that the straight line connecting the pair of A6 and B6 pixels is the X axis (the slope is 0 45) (π / 4), which is an angle with the central pixel, is matched to the center pixel.

The angular value matching unit 115 selects a symmetric pixel pair having a smaller difference in brightness value Y between the symmetric pixel pairs when there are a plurality of symmetric pixel pairs corresponding to the minimum value.

As described above, the central pixel matching unit 110 matches angular values with respect to a plurality of central pixels in the same manner as described above, and as a result, it can be expressed as shown in FIG.

7 is a view for explaining the step S320 of FIG.

The calculating unit 120 selects a center pixel pair having a minimum difference in angular value of a center pixel pair symmetrical about a target pixel located between a plurality of center pixels whose angular values are matched, The average of the values is calculated (S320).

Here, the target pixel is a pixel included in the interpolation pixel, and the target pixel to be interpolated in accordance with the embodiment of the present invention is referred to as a target pixel for convenience.

For example, in FIG. 7A, when there are two center pixel pairs (A1, B1) and (A2, B2) for the target pixel I (c) It is assumed that the angular value of B1 is? / 3, the angular value of A2 is? / 4, and the angular value of B2 is? / 6.

Then, the difference between the angular values of A1 and B1 symmetrical with respect to the target pixel I (c) is (π / 2 - π / 3 = π / 6) and A2 and B2 symmetrical with respect to I (? / 4 -? / 6 =? / 12), the calculation unit 120 selects the center pixel pair of A2 and B2 having the minimum difference in angle value. Then, the calculation unit 120 calculates ((? / 4 +? / 2) = 5? / 24) which is an average of the angular values of the selected center pixel pair and matches with the calculated average value. Here, the number of center pixel pairs may vary depending on the position of the center pixel. Fig. 7 (b) shows another example of the case where there are three central pixel images.

On the other hand, when there are a plurality of pairs of central pixels having the minimum angle value, the calculating unit 120 selects a center pixel pair having a minimum difference in brightness value (Y) between the center pixel pairs, To match the target pixel.

7A, the difference between the center pixel pairs of A1 and B1 and the center pixel pairs of A2 and B2 are the same, the brightness value Y of A1 is 0.5, the brightness value of B1 (Y) is 0.6, the brightness value (Y) of A2 is 0.6, and the brightness value (Y) of B2 is 0.4.

Then, the difference between the brightness values (Y) of the central pixel pairs of A1 and B1 where the difference in brightness value (Y) between the center pixel pairs is (0.6 - 0.5 = 0.1) 0.2), the arithmetic unit 120 selects the center pixel pair of A1 and B1, calculates the average angular value of A1 and B1, and matches the target pixel I (c).

The control unit 130 extends a virtual straight line corresponding to the average angle value matched around the target pixel and determines the brightness of the two intersecting pixels where the extended straight line crosses a virtual straight line between neighboring adjacent reference pixels Values (Y), Cr, and Cb are obtained (S330).

FIG. 8 is a schematic diagram of pixels necessary for explaining the step S330 of FIG. 3. FIG.

For example, in FIG. 8A, when the target pixel is located among four center pixels and the average angle value of the center pixel pair selected in step S320 is 3? / 8, Which is a value of 3? / 8.

In Fig. 8 (a), two intersecting pixels are selected at a point where an extended straight line (arrow) crosses two parallel imaginary straight lines between adjacent adjacent center pixels (indicated by the circles of dashed lines). The controller 130 obtains the brightness values (Y), Cr, and Cb values of the two intersecting pixels.

Here, when two intersecting pixels that intersect with a virtual straight line are not reference pixels, that is, when a pixel value is not set, the linear interpolator 113 calculates the brightness value (represented by a circle of a broken line) (Y), Cr, and Cb values of two intersecting pixels are linearly interpolated using the values of Y, Cr, and Cb.

When the average angle value of the center pixel pair corresponds to? / 4, the two intersecting pixels intersecting the virtual straight line become the reference pixel. In this case, the controller 130 calculates the brightness value Y , The Cb value is selected as the brightness values (Y), Cr, and Cb values of the two intersecting pixels.

Another form in which the two intersecting pixels are generated is shown in Fig. 8 (b) when the form of selecting the center pixel differs according to various embodiments of the present invention.

Finally, the setting unit 140 sets the brightness values Y, Cr, and Cb of the target pixel by averaging the brightness values Y, Cr, and Cb of the two intersecting pixels at step S340.

In addition, the conversion unit (not shown) may convert the brightness values (Y), Cr, and Cb of the target pixel into R, G, and B values (S350). The method of converting the YCrCb into the RGB form can be easily carried out by those skilled in the art, and thus a detailed description thereof will be omitted.

FIG. 9 is an exemplary view showing an original image before image enlargement. FIG. 10 is an enlarged view of an original image of FIG. 9 enlarged by an existing method.

In the case of FIG. 10, it can be seen that the phenomenon in which the pixels are crushed due to noise in the vicinity of the edge between the different colors or the boundary between the object and the background is serious. This occurs noticeably at pixel locations with edges of the order of the pixel thickness.

11 is a view showing an image obtained by enlarging an original image of FIG. 9 by a method according to an embodiment of the present invention.

In contrast, in the case of FIG. 11, in the case of enlarging an image by the method according to the present invention at a pixel position with an edge of about pixel thickness, the angle value according to the present invention ), The jagging phenomenon, which is a relatively stepped phenomenon, is reduced and the noise is reduced. Judging from the human visual standpoint, the color of the image can be prevented from being crushed by the present invention.

As described above, according to the apparatus and method for estimating a pixel value of a target pixel according to an embodiment of the present invention, by generating an interpolation pixel closest to an actual edge pixel in the vicinity of an edge of an image, noise due to image enlargement is removed, Can be prevented.

In addition, it is possible to alleviate the jagging phenomenon existing in the existing technology.

The present invention has been described above with reference to the embodiments. 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 as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. Therefore, the scope of the present invention is not limited to the above-described embodiments, but should be construed to include various embodiments within the scope of the claims and equivalents thereof.

Meanwhile, the embodiments of the present invention can be embodied as computer readable codes on a computer readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored.

Examples of the computer-readable recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device and the like, and also a carrier wave (for example, transmission via the Internet) . In addition, the computer-readable recording medium may be distributed over network-connected computer systems so that computer readable codes can be stored and executed in a distributed manner. And functional programs, codes, and code segments for implementing the present invention can be easily deduced by programmers skilled in the art to which the present invention pertains.

100: a pixel value estimation device of a target pixel
110: center pixel matching unit, 111: virtual processing unit,
112: cross pixel selection unit, 113: linear interpolation unit,
114: Summing unit, 115: Angle value matching unit,
120: operation unit, 130: control unit,
140: Setting section

Claims (12)

A method of estimating a pixel value of a target pixel located between a plurality of reference pixels with pixel values set,
Wherein a sum of difference values of the center pixel and a brightness value (Y) of the center pixel among the N pairs of symmetric pixels divided at the same angle around the center pixel, which is any one of the plurality of reference pixels, Selecting an asymmetrical pixel pair and matching an angular value between a virtual straight line connecting the selected symmetric pixel pair and the reference axis to the center pixel;
A center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel positioned between a plurality of center pixels having the angular values matched is selected, Selecting a center pixel pair having a minimum difference in brightness value (Y) between the center pixel pairs and calculating an average of angular values of the selected center pixel pair when the center pixel pair is a plurality;
(Y) of two intersecting pixels in which an imaginary straight line corresponding to the average angular value is extended about the target pixel and the extended straight line intersects with a virtual straight line between adjacent adjacent reference pixels, Obtaining a Cr, Cb value; And
And setting the brightness values (Y), Cr, and Cb values of the target pixels by averaging the brightness values (Y), Cr, and Cb values of the two intersecting pixels. The method according to claim 1,
Wherein the plurality of reference pixels are spaced apart at equal intervals. 3. The method of claim 2,
Wherein matching the angular value with the selected symmetric pixel pair to the center pixel comprises:
Covering a virtual window of a rectangular shape such that the reference pixel is positioned at each corner around the center pixel;
Generating N virtual straight lines passing through the center pixel and evenly arranged at the same angle, selecting an intersecting pixel located at an intersection of the imaginary straight line and the virtual window;
Linearly interpolating a brightness value (Y) of the crossing pixel using a brightness value (Y) of a surrounding reference pixel when the crossing pixel is not the reference pixel;
Calculating a difference value between a first crossing pixel included in the N symmetric pixel pairs and a brightness value (Y) of the center pixel and a difference value between the remaining second crossing pixel and a brightness value (Y) of the center pixel, ; And
Selecting a minimum value among the N values generated by the summing and matching an angle value between a reference line and a virtual line connecting a pair of symmetric pixels corresponding to the selected minimum value to the center pixel; / RTI > delete 3. The method of claim 2,
The step of obtaining the brightness values (Y), Cr, and Cb values of the two intersecting cross-
(Y), Cr, and Cb values of the cross pixels using the brightness values (Y), Cr, and Cb values of surrounding reference pixels when the two intersecting pixels intersecting the virtual straight line are not the reference pixels ≪ / RTI >
(Y), Cr, and Cb values of the reference pixel as the brightness values (Y), Cr, and Cb values of the two intersecting pixels when the two intersecting pixels are the reference pixels, How to estimate. 3. The method of claim 2,
And converting the brightness values (Y), Cr, and Cb values of the target pixel to R, G, and B values. An apparatus for estimating a pixel value of a target pixel located between a plurality of reference pixels having pixel values set therein,
Wherein a sum of difference values of the center pixel and a brightness value (Y) of the center pixel among the N pairs of symmetric pixels divided at the same angle around the center pixel, which is any one of the plurality of reference pixels, A center pixel matching unit for selecting an asymmetrical pixel pair and matching an angular value between a virtual straight line connecting the selected symmetric pixel pair and the reference axis to the center pixel;
A center pixel pair having a minimum difference in angular value of a center pixel pair symmetric about the target pixel positioned between a plurality of center pixels having the angular values matched is selected, An operation unit for selecting a center pixel pair having a minimum difference in brightness value Y between the center pixel pairs and calculating an average of the angular values of the selected center pixel pair when the center pixel pairs are plural;
(Y) of two intersecting pixels in which an imaginary straight line corresponding to the average angular value is extended about the target pixel and the extended straight line intersects with a virtual straight line between adjacent adjacent reference pixels, Cr, and Cb; And
And a setting unit for setting the brightness values (Y), Cr, and Cb values of the target pixel by averaging the brightness values (Y), Cr, and Cb values of the two intersecting pixels. 8. The method of claim 7,
Wherein the plurality of reference pixels are spaced at equal intervals. 9. The method of claim 8,
Wherein the central pixel matching unit comprises:
A virtual processing unit for covering a virtual window of a rectangular shape such that the reference pixel is positioned at each corner around the center pixel;
A cross pixel selecting unit for generating N virtual straight lines passing through the center pixel and evenly arranged at the same angle and selecting an intersecting pixel located at an intersection of the virtual straight line and the virtual window;
A linear interpolator that linearly interpolates the brightness value Y of the crossing pixel using the brightness value Y of the surrounding reference pixel when the crossing pixel is not the reference pixel;
(Y) of the center pixel and the difference value of the remaining second crossing pixel and the brightness value (Y) of the center pixel are added to each other, the sum of the difference values of the first crossing pixel included in the N symmetric pixel pairs and the brightness value part; And
And an angular value matching unit for selecting an angular value between the imaginary straight line connecting the symmetric pixel pair corresponding to the selected minimum value and the reference axis to the center pixel, Pixel value estimator. delete 9. The method of claim 8,
Wherein,
(Y), Cr, and Cb values of the cross pixels using the brightness values (Y), Cr, and Cb values of surrounding reference pixels when the two intersecting pixels intersecting the virtual straight line are not the reference pixels ≪ / RTI >
(Y), Cr, and Cb values of the reference pixel as the brightness values (Y), Cr, and Cb values of the two intersecting pixels when the two intersecting pixels are the reference pixels, Device. 9. The method of claim 8,
And an RGB converter for converting the brightness values (Y), Cr, and Cb values of the target pixel into R, G, and B values.

Images