KR101523767B1 - Apparatus and method for encoding/decoding color image - Google Patents

Abstract

The present invention relates to a color image encoding / decoding apparatus, which comprises: an image sensing unit for acquiring a sampled color image of a first color coordinate system; And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels by performing linear prediction at a specific position, A color conversion unit for calculating a component value of a second color coordinate system at a specific position and converting the calculated component value into a color image of the second color coordinate system based on the calculated result; And an encoding unit encoding the converted color image of the second color coordinate system to generate a compressed image.

Description

[0001] APPARATUS AND METHOD FOR ENCODING / DECODING COLOR IMAGE [0002]

The present invention relates to a color image encoding / decoding apparatus and method, and more particularly, to an apparatus and method for encoding / decoding a sampled color image obtained using a CFA (Color Filter Array).

Most video devices such as digital cameras, digital camcorders, and portable camera phones store images in a digital format. For example, most digital cameras compress images using compression techniques such as JPEG (Joint Photographic Expert Group). In addition, many digital camcorders also store moving pictures in digital format.

More and more video devices store still images and movies in digital format. Most mobile phones are equipped with cameras and can transmit still images and moving images via digital communication.

The color image sampled in an image system using a single image sensor with CFA is interpolated into a full color image by an interpolation algorithm. The data amount of the full color image generated by the interpolation process is three times the data amount of the sampled color image.

FIG. 1 is a first diagram for explaining a principle of converting a color image according to the related art to solve such a problem, and FIG. 2 is a second diagram for explaining the principle of converting a color image according to the related art.

Referring to FIG. 1, in order to solve such a problem, a method has been proposed in which the obtained RGB color image is converted into a color image of a second color coordinate system such as Yuv by a predetermined number of pixels.

At this time, Y1, Y2, u, and v of the Yuv color coordinate system can be calculated as shown in the following equation (1).

[Equation 1]

_{Y1 = w 11 * R + w} 12 * G1 + w 13 * B

_{Y2 = w 21 * R + w} 22 * G2 + w 23 * B

u = w ₃₁ * R + w ₃₂ * G 1 + w ₃₂ * G 2 + w ₃₃ * B

v = w ₄₁ * R + w ₄₂ * G 1 + w ₄₂ * G 2 + w ₄₃ * B.

In addition, the following Equation (2) can be used by modifying Equation (1).

&Quot; (2) "

_{Y1 = w 11 * R + w} 12 * G1 + w 13 * B

_{Y2 = w 21 * R + w} 22 * G2 + w 23 * B

u = w ₃₁ * R + w ₃₂ * G 1 + w ₃₃ * B

v = w ₄₁ * R + w ₄₂ * G2 + w ₄₃ * B

In calculating the Y1, Y2, u, and v component values of the Yuv color coordinate system using G1, G2, R, and B located at different positions in the above Equation 2, high frequency components have.

Referring to FIG. 2, when G1 and G2 located on different horizontal axes are converted to Y1 and Y2 of the same horizontal axis, high frequency components that are not present in the original image may be generated, which may result in lower coding efficiency .

Also, when the G1 and G2 values located on the other horizontal axes are converted to Y1 and Y2 of the same horizontal axis, errors may occur in motion estimation and motion compensation in moving picture encoding, which makes it difficult to obtain an optimal solution .

Therefore, in consideration of the fact that the RGB pixels of the color image of the RGB color coordinate system acquired by using the CFA (Color Filter Array) exist at different positions, the RGB Decoding the color image obtained by using the CFA that linearly predicts the color image and converts the color image into a color image of a different color coordinate system and then encodes the color image, and a method of encoding / decoding the sampled color image.

It is another object of the present invention to provide a method and an apparatus for dividing an image into a plurality of regions in order to prevent inversion in case of performing linear prediction on RGB values corresponding to desired positions considering that RGB pixels exist at different positions, The present invention also provides an apparatus and method for encoding / decoding a sampled color image using a CFA that performs linear prediction only in each region and converts the color image into a color image of a different color coordinate system.

However, the objects of the present invention are not limited to those mentioned above, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided an apparatus for encoding a color image, comprising: an image sensing unit for acquiring a sampled color image of a first color coordinate system; And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels by performing linear prediction at a specific position, A color conversion unit for calculating a component value of a second color coordinate system at a specific position and converting the calculated component value into a color image of the second color coordinate system based on the calculated result; And an encoding unit encoding the converted color image of the second color coordinate system to generate a compressed image.

Preferably, the first color coordinate system is an RGB color coordinate system.

Preferably, the second color coordinate system is a YUV color coordinate system.

Preferably, the color conversion unit converts the sampled color image of the first color coordinate system into a color image of the second color coordinate system as a unit of a plurality of pixels.

According to another aspect of the present invention, there is provided an apparatus for encoding a color image, comprising: an image sensing unit for acquiring a sampled color image of a first color coordinate system; The color image sampled in the first color coordinate system is converted into a color image in the second color coordinate system having the same number of pixels. In this case, the entire image is divided into a plurality of regions, and then the component values of the first coordinate system A color conversion unit for calculating a component value of the second color coordinate system and converting the calculated component value into a color image of the second color coordinate system based on the calculated result; And an encoding unit encoding the converted color image of the second color coordinate system to generate a compressed image.

Preferably, the color converter divides the entire image into a plurality of regions and then performs a linear prediction at the specific position using only pixels of a region including the specific position at a specific position, And calculates a component value and a component value of a second color coordinate system of the specific position.

In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels by using the CFA according to another aspect of the present invention, And a component value of a second color coordinate system at the specific position, and decodes a signal obtained by compressing an image obtained by converting the color component of the second color coordinate system into a color image based on the calculated result, The apparatus comprising: a decoding unit decoding a received compressed signal to generate a decoded color signal of a second color coordinate system; And a color inversion unit for performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system.

In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels by using the CFA according to another aspect of the present invention, When calculating the component values of the second color coordinate system, the component values of the first coordinate system and the second color coordinate system of the specific position are calculated using only the pixels of the region including the specific position, The decoding apparatus decodes a signal obtained by compressing an image obtained by performing a calculation on the basis of a result of the calculation and transforming the image into the color image of the second color coordinate system. The decoding apparatus decodes the received compressed signal to generate a decoded color signal of the second color coordinate system ; And a color inversion unit for performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system.

According to another aspect of the present invention, there is provided a method of encoding a color image, the method comprising: sensing an image of a color sampled in a first color coordinate system; And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels by performing linear prediction at a specific position, A color conversion step of calculating a component value of a second color coordinate system at a specific position and converting the component value into a color image of the second color coordinate system based on the calculated result; And encoding the converted color image of the second color coordinate system to generate a compressed image.

Preferably, the first color coordinate system is an RGB color coordinate system.

Preferably, the second color coordinate system is a YUV color coordinate system.

Preferably, the color conversion step converts the sampled color image of the first color coordinate system into a color image of the second color coordinate system as a unit of a plurality of pixels.

According to another aspect of the present invention, there is provided an apparatus for encoding a color image, comprising: an image sensing step of acquiring a sampled color image of a first color coordinate system; The color image sampled in the first color coordinate system is converted into a color image in the second color coordinate system having the same number of pixels. In this case, the entire image is divided into a plurality of regions, and then the component values of the first coordinate system A color conversion step of calculating a component value of a second color coordinate system and converting the result into a color image of the second color coordinate system based on the calculated result; And encoding the converted color image of the second color coordinate system to generate a compressed image.

Preferably, the color conversion step may include a step of dividing the entire image into a plurality of regions, performing a linear prediction at the specific position using only pixels of a region including the specific position at a specific position, And a component value of a second color coordinate system of the specific position.

In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels by using the CFA according to another aspect of the present invention, And a component value of a second color coordinate system at the specific position, and decodes a signal obtained by compressing an image obtained by converting the color component of the second color coordinate system into a color image based on the calculated result, A decoding step of decoding a received compressed signal to generate a decoded color signal of a second color coordinate system; And a color inversion step of performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system.

In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels by using the CFA according to another aspect of the present invention, When calculating the component values of the second color coordinate system, the component values of the first coordinate system and the second color coordinate system of the specific position are calculated using only the pixels of the region including the specific position, And decodes a signal obtained by compressing the image converted into the color image of the second color coordinate system based on the calculated result, the decoding method decodes the received compressed signal to generate a decoded color signal of the second color coordinate system ; And a color inversion step of performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system.

In consideration of the fact that the RGB pixels of the color image of the RGB color coordinate system obtained using the CFA exist at different positions, the present invention linearly predicts the RGB values corresponding to the desired positions, And then encodes it, it is possible to effectively encode and decode the color image.

1 is a first diagram for explaining a principle of converting a color image according to the related art.
2 is a second diagram for explaining the principle of converting a color image according to the related art.
3 is a block diagram of a color image encoding apparatus according to an embodiment of the present invention.
4 is a diagram showing a CFA of a Bayer pattern according to an embodiment of the present invention.
5 is a linear interpolation matrix for explaining a color image encoding process according to an embodiment of the present invention.
6 is a backward linear interpolation matrix for explaining a color image decoding process according to an embodiment of the present invention.
7 is a backward linear interpolation matrix of another example for explaining a color image decoding process according to an embodiment of the present invention.
FIG. 8 is a diagram illustrating an example of linear prediction after dividing an image in a horizontal direction in a color image encoding process according to an exemplary embodiment of the present invention. Referring to FIG.
9 is a diagram illustrating an example in which an image is divided into small blocks and linearly predicted in a color image encoding process according to an embodiment of the present invention.
10 is a diagram illustrating an example of converting an RGB image sampled without interpolation into a Yuv image.
11 is a diagram illustrating a color image encoding method according to an embodiment of the present invention.
12 is a block diagram of a color image decoding apparatus according to an embodiment of the present invention.
13 is a diagram illustrating a color image decoding method according to an embodiment of the present invention.

Hereinafter, a color image encoding / decoding apparatus and method according to an embodiment of the present invention will be described with reference to the accompanying drawings. The present invention will be described in detail with reference to the portions necessary for understanding the operation and operation according to the present invention.

In describing the constituent elements of the present invention, the same reference numerals may be given to constituent elements having the same name, and the same reference numerals may be given thereto even though they are different from each other. However, even in such a case, it does not mean that the corresponding component has different functions according to the embodiment, or does not mean that the different components have the same function. It should be judged based on the description of each component in the example.

In particular, in consideration of the fact that the RGB pixels of the color image of the RGB color coordinate system obtained by using the CFA (Color Filter Array) exist at different positions, the RGB values corresponding to the desired positions are linearly predicted, We propose a new color image coding method that converts the color image into a color image of the coordinate system and then encodes it.

In other words, a color image sampled in an image system using a single image sensor having CFA is interpolated into a full color image by an interpolation algorithm. The data amount of the RGB full color image generated by this interpolation process is three times the data amount of the sampled color image. Even if this is converted to the Yuv420 format, the data amount is increased by 1.5 times and the coding efficiency is lowered. To solve this problem, RGB images are converted into Yuv images having the same number of pixels. In the conventional method, RGB pixels are converted into Yuv pixels based on a predetermined number of pixels (for example, four). At this time, the values of G1 and G2 located on the other horizontal axis are converted into Y1 and Y2 of the same horizontal axis, so that high frequency components that are not present in the original image may be generated, thereby causing a reduction in coding efficiency. Also, since the R, G, and B component values at different positions are used to calculate the Y, u, and v values, high frequency components that did not exist in the original image may be additionally generated. In order to solve this problem, in the present invention, the RGB values corresponding to the desired positions are linearly predicted, converted into Yuv color images, and then encoded.

3 is a block diagram of a color image encoding apparatus according to an embodiment of the present invention.

3, the color image encoding apparatus 300 according to the present invention may include an image sensing unit 310, a color conversion unit 320, an encoding unit 330, and the like.

The image sensing unit 310 may acquire a sampled color image of the first color coordinate system using the CFA.

The color conversion unit 320 may convert the sampled color image of the obtained first color coordinate system into a color image of the second color coordinate system having the same number of pixels. Here, the first color coordinate system represents the RGB color coordinate system, and the second color coordinate system represents the Yuv color coordinate system.

At this time, when calculating the component value of the second color coordinate system at the specific position, the color conversion unit 320 performs the linear prediction to correct the positional difference to calculate the component value of the first coordinate system of the specific position and the second color coordinate system The color component of the first color coordinate system can be converted into the color image of the second color coordinate system based on the calculated result.

In addition, the color converter 320 divides the entire image into a plurality of areas, calculates a component value of the second color coordinate system at a specific position, and performs linear prediction using only pixels in the corresponding area to obtain a first coordinate system And a component value of a second color coordinate system at a specific position is calculated and converted into a color image of the second color coordinate system based on the calculated result.

The encoding unit 330 may encode the color image of the second color coordinate system to generate a compressed image.

4 is a diagram showing a CFA of a Bayer pattern according to an embodiment of the present invention.

Referring to FIG. 4, 100 pixels of a 10 × 10 image sensor are numbered from 1 to 100. FIG. 3 shows a CFA Bayer pattern consisting of 50% green and 25% blue and red. In order to maintain the coding efficiency, when converting the image of FIG. 3 into Yuv coordinates, the Y pixel should be converted to 50%, and the u and v pixels should be converted to 25%.

Here, when converting 100 RGB pixels to 100 Yuv pixels, the conversion can be expressed as shown in FIG. 10 which will be described later. The conversion relation can be expressed by the following equation (3).

&Quot; (3) "

In the above equation (3), A is a matrix of 100 x 100, and 100 RGB pixels are represented by a vector (100 x 1). In Equation (3), the subscript denotes the pixel position.

Likewise, 100 Yuv values are also represented by a vector (100 x 1). If there is an inverse matrix of the matrix A, the original RGB value can be obtained as shown in the following equation (4).

&Quot; (4) "

The equation (4) can be expressed as the following equation (5).

&Quot; (5) "

V _RGB = A ^-1 V _Yuv where,

At this time, a row of the matrix A or the inverse matrix represents a weight for each pixel of the original image. That is, each row of the matrix A represents a weight for each pixel of a sampled RGB image required to calculate a Y or u or v component value at a specific position, and a row of an inverse matrix represents a weight of a specific position of the sampled RGB image Represents the weight for each pixel of the transformed Yuv image that requires R, G, or B component values.

From the viewpoint of the encoder, Y1 and Y2 must exist on the same horizontal axis. Also, the red and blue values corresponding to pixel 1 should be used in calculating the Y1 value.

However, in calculating the Y component value of the pixel 24 in the previous method, the Y value of the pixel 24 is calculated using the green value of the pixel 34, the blue value of the pixel 33, and the red value of the pixel 24, Components are generated to lower the coding efficiency. To solve this problem, in the present invention, the green and blue values of the pixel 24 are linearly predicted to calculate the Y value at the pixel 24.

The blue value of the pixel 24 in the middle area can be linearly predicted with the following equation (6).

&Quot; (6) "

Further, the green value of the pixel 24 can be calculated by the following equation (7).

&Quot; (7) "

Assuming that Y = w _yr * R + w _yg * G + w _yb * B, the 34th row of the matrix A corresponding to the pixel 34 has the same value as the following Equation 8 .

&Quot; (8) "

a _{34 , 24} = w _yr

a _{34 , 13} = a _{34 , 15} = a _{34 , 33} = a _{34 , 35} = w _yb / 4

a _{34 , 23} = a _{34 , 25} = w _yg / 2

Here, a _{i , j} means the elements of the i-th row and the j-th column of the matrix A, and the values of the remaining elements of the 34-th row are all zero.

It is also possible to calculate the green value as in the following Equation (9), and it is of course possible to calculate the R, G and B values at specific positions using other linear prediction methods.

&Quot; (9) "

, 또는

, or

In a similar manner, the red value in the pixel 25 can be calculated as shown in the following equation (10).

&Quot; (10) "

Further, the blue value in the pixel 25 can be obtained by the following equation (11).

&Quot; (11) "

At this time, the 25th row of the matrix A has the same value as the following Equation (12).

&Quot; (12) "

a _{25 , 25} = w _yg

a _{25 , 24} = a _{25 , 26} = w _yr / 2

a _{25 , 15} = a _{25 , 35} = w _yb / 2

Here, the values of the remaining elements of the 25th row are all zero.

In this way, the values of Y, u, and v in the intermediate region can be obtained in a similar manner, and the row of the corresponding matrix A can also be determined. Boundary rows or columns (eg, 1 row or 1 column) must predict the green, red, and blue values of the pixel in a different way.

For example, the values of R1 and B1 in pixel 1 can be linearly predicted as shown in the following equation (13). Of course, not only these functions but also other linear prediction functions can be used.

&Quot; (13) "

Therefore, the Y value in pixel 1 can be calculated as a linear sum of G1, R1, and B1.

Also, the green pixel value G2 at the position of the pixel 2 can be found by the following equation (14).

&Quot; (14) "

Further, the green pixel value G2 can be obtained by using a line format such as the following formula (15).

&Quot; (15) "

Similarly, the blue value at the position of the pixel 2 can be obtained by the following equation (16).

&Quot; (16) "

From Equation (16) thus obtained, the following Equation (17) can also be approximated.

&Quot; (17) "

In order to obtain the v value at the pixel 24 at another position, the green value and the blue value in the pixel 24 can be obtained as shown in the following equations (18) and (19), respectively.

&Quot; (18) "

&Quot; (19) "

Therefore, the value of v24 can be calculated using the values of R24, G24, and B24.

Assuming v = w _vr * R + w _vg * G + w _vb * B, the 24th row of the matrix A corresponding to the pixel 24 has the same value as in the following Equation (20).

&Quot; (20) "

A _24,24 = w _vr

_{A 24,13 = a 24, 15 =} a 24, 33 = a 24, 35 = w vb / 4

_{A 24,23 = a 24, 25 =} w vg / 2

The u value can be calculated in pixel 33 in a similar manner.

The Yuv vector (V _Yuv ) can be obtained by multiplying the RGB vector (V _RGB ) by the matrix A generated in this way, and the Yuv image can be obtained from this Yuv vector. In this Yuv image, the high frequency component that lowers the coding efficiency can be remarkably reduced. In general, the number of pixels of a moving picture has a very large value. For example, in the case of HDTV, there are about 200 million pixels, and in this case matrix A becomes 2,000,000 x 2,000,000, making it almost impossible to calculate the inverse matrix. However, in performing the linear prediction, the linear prediction function can be selected so that a section having a non-zero value locally occurs in the inverse matrix of the matrix A and the matrix A. In Equation (5), the vectors V _RGB and V _Yuv represent the entire image, and the row of the matrix A or the inverse matrix represents a weight for each pixel of the original image. That is, each row of the matrix A represents a weight for each pixel of the sampled RGB image required to calculate a Y or u or v component value at a specific position, and a row of the inverse matrix represents R Or a weight value for each pixel of the converted Yuv image that requires a G or B component value. Therefore, if a row of a matrix A or an inverse matrix is displayed in a circular shape, a weight necessary for calculating the corresponding pixel can be schematically shown.

In the present invention, a linear interpolation method is selected so that a non-zero section of a row of a matrix A or an inverse matrix occurs locally. If the linear interpolation method is selected as described above, the same pattern is repeatedly generated in the intermediate region, so that only the non-zero region can be stored and the color coordinate change can be efficiently performed. For example, a row of the matrix A used to calculate a Yuv component at a specific position may be displayed in the form of an original image to have a pattern as shown in FIG. 4 (only partial display, zero or a section having a very small value is omitted). Since the same pattern is repeatedly generated in the intermediate region, the color conversion operation (RGB-> Yuv) can be efficiently performed using a small amount of memory.

If the corresponding inverse matrix row is displayed again in the form of an original image, it is given as shown in FIG. 5 (partial display only, zero or a section having a very small value is omitted).

That is, FIG. 6 shows an example where a non-zero period occurs locally in the row of the inverse matrix. Likewise, since the same pattern is repeatedly generated in the intermediate region, the color conversion operation (Yuv- > RGB) can be efficiently performed using a small capacity memory.

That is, since the same pattern is repeatedly generated in the intermediate region, a retrograde column can be obtained using a constant size matrix (for example, a matrix of about 40 × 40) that can be implemented and applied to an HDTV-sized image. In the case of a boundary row or a column, it should be processed accordingly. It goes without saying that the matrix size can be changed according to the linear function used.

Figure 7 is an example where a non-zero interval in the row of the inverse matrix occurs in almost all regions. If the non-zero interval does not occur locally as shown in FIG. 7, the linear prediction function should be adjusted. In general, the inverse filter of the FIR filter appears as an IIR filter. Therefore, theoretically, the non-zero region occupies the entire image, but most values can be made very small, so that it is possible to reduce the residual error to a negligible extent even if approximated by an FIR filter.

In another embodiment of the present invention, when linear prediction is performed in order to prevent non-zero intervals from occupying the whole image of FIG. 7 (i.e., elements of rows of an inverse matrix are almost non-zeal) In predicting the pixel values of each divided area by dividing the entire image, only the pixels belonging to the same area are used for prediction. For example, in FIG. 8, when an image is divided into a plurality of horizontal strips and pixel values within a specific strip are predicted, linear prediction is performed using only pixels of the same strip. In this case, the non-zero period of the inverse matrix row can be locally limited. It goes without saying that the strip can be constructed in the longitudinal direction.

9 shows an example in which only pixels of the same block are used in the case of dividing an image into a plurality of blocks and predicting pixel values in the respective blocks. In this way, a more complex and sophisticated prediction function can be used when dividing an image and applying a linear prediction function. In general, when a complex linear prediction function is used, a non-zero interval of an inverse matrix row occupies the entire image, making it impossible to apply it to a large-screen field such as an HDTV.

11 is a diagram illustrating a color image encoding method according to an embodiment of the present invention.

As shown in FIG. 11, the color image encoding apparatus according to the present invention can acquire a sampled color image of the first color coordinate system using CFA (S1110).

Next, when calculating the component values of the second color coordinate system at a specific position, the color image coding apparatus performs linear prediction to compensate for the positional difference, thereby calculating the component values of the first coordinate system at a specific position and the second color coordinate system Component values are calculated and converted into a color image of the second color coordinate system based on the calculated result (S1120).

The color image coding apparatus divides the entire image into a plurality of areas and calculates a component value of a second color coordinate system at a specific position. The color image coding apparatus performs linear prediction using only pixels in the corresponding area to calculate a component value of a first coordinate system And a component value of a second color coordinate system at a specific position is calculated and converted into a color image of the second color coordinate system based on the calculated result.

Next, the color image encoding apparatus can generate a compressed image by encoding the converted color image of the second color coordinate system (S1130).

12 is a block diagram of a color image decoding apparatus according to an embodiment of the present invention.

12, the color image decoding apparatus according to the present invention may include a decoding unit 1210, a color inversion unit 1220, a color interpolation unit 1230, and the like.

The decoding unit 1210 may generate the decoded color signal of the second color coordinate system by decoding the received compressed signal.

The color inverse transform unit 1220 may generate an inverse decoded color signal of the first color coordinate system by performing an inverse conversion process in consideration of the linear prediction performed to compensate the positional difference of the decoded color signal of the second color coordinate system .

The color interpolator 1230 may interpolate the sampled decoded color signal of the generated first color coordinate system to generate a full resolution decoded color signal of the first color coordinate system.

13 is a diagram illustrating a color image decoding method according to an embodiment of the present invention.

As shown in FIG. 13, the color image decoding apparatus according to the present invention decodes the encoded Yuv image using a locally non-zero section of the row of the inverse transform matrix (S1310) Color coordinate conversion is applied (S1320). That is, an RGB image of a CFA pattern can be obtained by locally applying an inverse transformation matrix. Since the RGB image of the CFA pattern is the sampled RGB image, it is interpolated to generate a full resolution RGB image (S1330).

In the above embodiment, it is assumed that the position of u, v is different from the position of Y in calculating Y, u, v. However, in general, in the case of Yuv422 format image, u, v are calculated at the same position as Y. That is, in FIG. 4, in the case of a general Yuv422 format image, the values of u and v are calculated at pixel 1, pixel 3, pixel 43, The Y component value is calculated in pixel 1, pixel 2, pixel 3, pixel 4, and so on. That is, the Y, u, and v component values are calculated at the pixel 1 position, and only the Y component values at the pixel 2 position. Those skilled in the art will be able to calculate the u, v component values at the same location as Y according to the method presented in the present invention. For example, in order to obtain the values of Y, u, and v in the pixel 23, the red, green, and blue values of the pixel 23 are linearly predicted, and the values of Y, u, and v are obtained using the predicted values. In this case, it is needless to say that the red, green, and blue component values can be calculated at the same position and the Y, u, and v component values at the same position can be calculated in the same manner as described above.

It is to be understood that the present invention is not limited to these embodiments, and all of the elements constituting the embodiments of the present invention described above may be combined or operated in one operation. That is, within the scope of the present invention, all of the components may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. In addition, such a computer program may be stored in a computer-readable medium such as a USB memory, a CD disk, a flash memory, etc., and read and executed by a computer, thereby implementing embodiments of the present invention. As the storage medium of the computer program, a magnetic recording medium, an optical recording medium, a carrier wave medium, or the like may be included.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or essential characteristics thereof. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

310: a video sensing unit
320: color conversion unit
330:
1110:
1120: color inversion section
1130: Color interpolation part

Claims (16)

An image sensing unit for acquiring a sampled color image of the first color coordinate system;
And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels, performing linear prediction at a desired position to calculate a component value of the first coordinate system of the predetermined position A color conversion unit for calculating a component value of the second color coordinate system and converting the calculated component value into a color image of the second color coordinate system based on the calculated result; And
An encoding unit for encoding a color image of the second color coordinate system to generate a compressed image;
, And the color conversion unit
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, Converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system based on the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And converting the sampled color image of the first color coordinate system into a color image of the second color coordinate system. The method according to claim 1,
Wherein the first color coordinate system is an RGB color coordinate system. The method according to claim 1,
Wherein the second color coordinate system is a YUV color coordinate system. The method according to claim 1,
Wherein the color conversion unit comprises:
Wherein the sampled color image of the first color coordinate system is converted into a color image of the second color coordinate system as a unit of a plurality of pixels. An image sensing unit for acquiring a sampled color image of the first color coordinate system;
And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels, dividing the entire image into a plurality of areas, Of the second color coordinate system based on the result of the calculation of the component value of the first coordinate system and the second color coordinate system by performing the linear prediction using only the pixels of the second color coordinate system, A conversion unit; And
An encoding unit for encoding a color image of the second color coordinate system to generate a compressed image;
, And the color conversion unit
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, Converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system based on the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And converting the sampled color image of the first color coordinate system into a color image of the second color coordinate system. delete In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels using CFA, linear prediction is performed at a desired position to calculate a component value of the first coordinate system of the desired position A decoding apparatus for decoding a signal obtained by compressing an image obtained by calculating a component value of a second color coordinate system and converting the color value of the second color coordinate system into a color image based on the calculated result,
A decoding unit decoding the received compressed signal to generate a decoded color signal of a second color coordinate system; And
A color inversion unit for performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system;
, And the color image of the second color coordinate system
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, The sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system on the basis of the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system Wherein the sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system. In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels using CFA, the entire image is divided into a plurality of regions, and the component values of the second color coordinate system The component values of the first coordinate system and the second color coordinate system of the desired position are calculated using only the pixels of the region including the desired position, and based on the calculated results, And a decoding unit which decodes a signal obtained by compressing an image converted into a color image of the second color coordinate system,
A decoding unit decoding the received compressed signal to generate a decoded color signal of a second color coordinate system; And
A color inversion unit for performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system;
, And the color image of the second color coordinate system
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, The sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system on the basis of the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system Wherein the sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system. An image sensing step of acquiring a sampled color image of the first color coordinate system;
And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels, performing linear prediction at a desired position to calculate a component value of the first coordinate system of the desired position, A color conversion step of calculating a component value of the two-color coordinate system and converting the calculated value into a color image of the second color coordinate system; And
A coding step of coding the color image of the second color coordinate system to generate a compressed image;
Wherein the color conversion step comprises the steps of:
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, Converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system based on the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system. 10. The method of claim 9,
Wherein the first color coordinate system is an RGB color coordinate system. 10. The method of claim 9,
Wherein the second color coordinate system is a YUV color coordinate system. 10. The method of claim 9,
Wherein the color conversion step comprises:
Wherein the sampled color image of the first color coordinate system is converted into a color image of the second color coordinate system as a unit of a plurality of pixels. An image sensing step of acquiring a sampled color image of the first color coordinate system;
And converting the sampled color image of the acquired first color coordinate system into a color image of a second color coordinate system having the same number of pixels, dividing the entire image into a plurality of areas, Of the second color coordinate system based on the result of the calculation of the component value of the first coordinate system and the second color coordinate system by performing the linear prediction using only the pixels of the second color coordinate system, Conversion step; And
A coding step of coding the color image of the second color coordinate system to generate a compressed image;
Wherein the color conversion step comprises the steps of:
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, Converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system based on the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system. delete In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels using CFA, linear prediction is performed at a desired position to calculate a component value of the first coordinate system of the desired position A decoding method for calculating a component value of a second color coordinate system and decoding a signal obtained by compressing an image obtained by converting a color image of the second color coordinate system based on the calculated result,
A decoding step of decoding the received compressed signal to generate a decoded color signal of a second color coordinate system; And
A color inversion step of performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system;
, Wherein the color image of the second color coordinate system
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, The sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system on the basis of the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And the sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system. In converting the sampled color image of the first color coordinate system into the color image of the second color coordinate system having the same number of pixels using CFA, the entire image is divided into a plurality of regions, and the component values of the second color coordinate system The component values of the first coordinate system and the second color coordinate system of the desired position are calculated using only the pixels of the region including the desired position, and based on the calculated results, A decoding method for decoding a signal obtained by compressing an image converted into a color image of the second color coordinate system,
A decoding step of decoding the received compressed signal to generate a decoded color signal of a second color coordinate system; And
A color inversion step of performing a reverse conversion process on the decoded color signal of the second color coordinate system in consideration of the linear prediction to generate a sampled decoded color signal of the first color coordinate system;
, Wherein the color image of the second color coordinate system
Calculating a component value of the first coordinate system by linearly predicting the R, G, and B values at the desired position using the R, G, and B values adjacent to the corresponding position, and using the calculated component values of the first coordinate system, The sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system on the basis of the calculated component value of the second color coordinate system after calculating the component value of the second color coordinate system,
When the color image of the second color coordinate system obtained through the conversion process is inversely transformed into the color image of the first color coordinate system using an inverse transformation function, the inverse transformation function locally uses the pixel of the color image of the second color coordinate system And the sampled color image of the first color coordinate system is converted into the color image of the second color coordinate system.

Images