KR20080066521A - Method and apparatus for deblocking filtering - Google Patents

Abstract

A method and an apparatus for deblocking filtering are provided to enable a current image processing unit and/or a different image processing unit adjacent to the current image processing unit to perform the deblocking filtering in consideration of a case that a current image block is encoded using illumination compensation, thereby removing blocking artifacts generated newly by the illumination compensation. A method for deblocking filtering comprises the followings: comparing the difference value of illumination change of a current image processing unit with the difference value of illumination change of a different image processing unit adjacent to the current image processing unit to determine whether two difference values are different(410); setting filtering intensity so that BS(Boundary Strength) is equal to K1 to perform the deblocking filtering of a current block boundary if it is determined that the two difference values are different(420); and setting the filtering intensity so that Bs is equal to K2 to perform the deblocking filtering of the current block boundary if it is determined that the two difference values are not different(430).

Description

Deblocking filtering method and apparatus {Method and apparatus for deblocking filtering}

1A and 1B illustrate an apparatus for encoding and decoding an image according to the prior art.

2 is a flowchart illustrating a method for deblocking filtering an image according to the prior art.

3A and 3B illustrate an apparatus for encoding and decoding an image using illumination compensation according to an embodiment of the present invention.

4 is a flowchart illustrating a method of deblocking and filtering an image according to an embodiment of the present invention.

5 is a flowchart illustrating a method for deblocking filtering an image according to another embodiment of the present invention.

6 is a flowchart illustrating a method for deblocking filtering an image according to another embodiment of the present invention.

7 illustrates an image processing unit and a current block according to an embodiment of the present invention.

8 illustrates syntax of a slice header according to an embodiment of the present invention.

9 illustrates a deblocking unit according to an embodiment of the present invention.

The present invention relates to a method and apparatus for deblocking filtering of an image, and more particularly, to a method and apparatus for deblocking and filtering an image including a block encoded using illumination compensation.

In an image compression scheme such as MPEG-1, MPEG-2, MPEG-4 H.264 / MPEG-4 Advanced Video Coding (AVC), a picture is divided into image processing units, that is, macroblocks, to encode an image. Then, each macro block is encoded using inter prediction or intra prediction.

Since these image compression methods are decoded after compression coding in macroblock units, blocking artifacts occur in the reconstructed image. A discrete cosine transform is independently performed and quantized for each subblock (for example, 4 × 4 block) included in the macroblock. Blocking artifacts occur because loss of the original image occurs while converting and quantizing independently without considering correlation between neighboring blocks.

The deblocking filter improves the quality of the final reconstructed image by smoothing the boundary error of blocks generated in the block-based coding. 1A and 1B and 2 will be described in detail.

1A and 1B illustrate an apparatus for encoding and decoding an image according to the prior art. An apparatus for encoding an image by performing inter prediction according to the H.264 standard and an apparatus for decoding an inter predicted image are illustrated.

Referring to FIG. 1A, the motion estimation unit 110 generates a motion vector of a current block to be encoded with reference to reference picture (s) stored in the frame memory 122. The motion compensator 112 generates a prediction block that is a prediction value of the current block based on the motion vector generated by the motion estimation unit 110.

The generated prediction block is subtracted from the original current block to generate a residue, and the generated residue is discrete cosine transformed and quantized by the DCT and quantization unit 114. The entropy coder 116 entropy encodes the quantized residue. The encoded residue is inserted into the bitstream together with the motion vector and transmitted.

The quantized residue in the DCT and quantization unit 114 is dequantized in the inverse quantization and inverse DCT unit 118 for use in prediction of a picture to be encoded next, inverse discrete cosine transformed, and then reconstructed.

The reconstructed residue is added to the prediction block and stored in the frame memory 122, which is deblocked and filtered by the deblocking unit 120 before being stored in the frame memory 122. If the restored block and the block to which the prediction block is added are stored in the frame memory 122 as it is, since the aforementioned blocking artifacts occur, the deblocking unit 120 performs the deblocking filtering and then the frame memory 122. Store in A detailed description of performing deblocking filtering according to the related art will be described later with reference to FIG. 2.

FIG. 1B illustrates an apparatus for decoding an inter prediction coded block by FIG. 1A.

Referring to FIG. 1B, the motion compensator 128 searches for a reference picture stored in the frame memory 132 based on data about a motion vector included in the bitstream to generate a prediction block of the current block.

The entropy decoding unit 124 receives the encoded residue and performs entropy decoding. The entropy decoded residue is dequantized by inverse quantization and inverse DCT unit 126 and inverse discrete cosine transformed. The residue, which has been inversely discrete cosine transformed and reconstructed, is added to the prediction block generated by the motion compensator 128 and reconstructed into a block before encoding. As in the case of encoding, when the reconstructed block is stored in the frame memory 132 as it is, since the aforementioned blocking artifacts are generated, the deblocking unit 130 stores the reconstructed block in the frame memory 122 after the deblocking filtering is performed. .

2 is a flowchart illustrating a method of performing deblocking filtering according to the prior art. The deblocking filtering method performed by the deblocking unit 120 or 130 will be described using the deblocking filtering according to the H.264 standard as an example.

Referring to FIG. 2, in step 201, the deblocking unit 120 or 130 determines whether at least one of two adjacent blocks p and q to perform deblocking filtering is intra coded.

If it is determined in step 201 that at least one of the blocks p and q is intra coded, it is determined in step 202 whether the blocks p and q to be subjected to deblocking filtering are blocks located at a macro block boundary. If blocks p and q are blocks located at the macro block boundary, in step 206, the filtering strength of the deblocking filter is set to the strongest Bs (Boundary strength) = 4 and deblocking filtering is performed. If the blocks p and q to be subjected to the deblocking filtering are blocks not located at the macroblock boundary, the deblocking filtering is performed by setting the filtering strength to Bs = 3 in step 207.

If it is determined in step 201 that the blocks p and q are both blocks that are not intra coded, in step 203 it is determined whether at least one of the blocks p and q has an orthogonal transform coefficient, that is, a discrete cosine transform coefficient. If at least one of the blocks p and q has an orthogonal transform coefficient, deblocking filtering is performed by setting the filtering strength to Bs = 2 in step 208, and if the blocks p and q do not have orthogonal transform coefficients, the process proceeds to step 204.

If both blocks p and q have no orthogonal transform coefficients in step 203, it is determined in step 204 whether the reference frames of blocks p and q are different or the number of reference frames is different. If the reference frame or the number of reference frames is different, deblocking filtering is performed by setting the filtering strength to Bs = 1 in step 209.

If it is determined in step 204 that the number of reference frames and reference frames of blocks p and q are the same, it is determined whether there is a difference in the motion vectors of blocks p and q in step 205. If it is determined that there is a difference in the motion vector, the deblocking filtering is performed by setting the filtering strength to Bs = 1 in step 209. If it is determined that there is no difference in the motion vector, the filtering strength is set to Bs = 0 in step 210. Blocking filtering is not performed.

As described with reference to steps 201 to 210, the deblocking filtering method according to the prior art does not consider a case where two adjacent blocks p and q are encoded using illumination change compensation. The average value (DC value) of the current block is different from the case of coded using the lighting change compensation and the case coded without using the lighting change compensation, and this does not take this point into consideration despite blocking artifacts. .

Accordingly, there is a need for a method and apparatus capable of deblocking filtering in consideration of a case of encoding using lighting change compensation.

SUMMARY OF THE INVENTION An object of the present invention is to provide a method and apparatus for deblocking filtering of an image for overcoming the above disadvantages of the prior art, and to provide a computer readable recording medium having recorded thereon a program for executing the method.

The deblocking filtering method of an image according to the present invention for solving the technical problem is a difference value of illumination change of a current image processing unit including a current block and another image adjacent to the current image processing unit. Determining whether the illumination change difference value of the processing unit is different; And deblocking filtering the boundary of the current block by adjusting a filtering strength of the deblocking filter based on the determination result.

According to a more preferred embodiment of the present invention, the determining may include comparing an absolute value of a difference between an illumination change difference value of the current image processing unit and an illumination change difference value of the other image processing unit than a first threshold value. It further comprises the step.

According to a more preferred embodiment of the present invention, the comparing may include comparing the absolute value to a second threshold value less than the first threshold value, when the comparison result indicates that the absolute value is less than the first threshold value. It further comprises a step.

According to a more preferred embodiment of the present invention, the determining of the difference in the illumination change of the encoded image processing unit without using the illumination compensation compensation (illumination compensation) of the current image processing unit and the other image processing unit. Setting to '0'; And determining whether an illumination change difference value of the current image processing unit and an illumination change difference value of the other image processing unit are different based on the setting result.

An apparatus for deblocking filtering an image according to the present invention for solving the technical problem includes a difference value of illumination change of a current image processing unit including a current block and the current image processing unit. A control unit to determine whether a difference in illumination change value of another image processing unit adjacent to is different; And a filtering unit configured to deblock block filter the boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result.

According to a preferred embodiment of the present invention, the control unit compares the absolute value of the difference between the illumination change difference value of the current image processing unit and the illumination change difference value of the other image processing unit than the first threshold value. It features.

According to a more preferred embodiment of the present invention, if the absolute value is less than the first threshold value, the controller compares the absolute value with a second threshold value less than the first threshold value. It is done.

According to a more preferred embodiment of the present invention, the control unit is configured to set the difference in the illumination change value of the image processing unit encoded without using illumination compensation among the current image processing unit and the other image processing unit to '0'. Set to ', and determine whether a difference in illumination change of the current image processing unit and a difference in illumination change of the other image processing unit are different based on the setting result.

According to a more preferred embodiment of the present invention, the image processing unit is characterized in that the macro block.

In order to solve the above technical problem, the present invention provides a computer-readable recording medium having recorded thereon a program for executing the above-described deblocking filtering method.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

3A and 3B illustrate an apparatus for encoding and decoding an image according to an embodiment of the present invention. An apparatus for encoding and decoding an image by performing inter prediction using illumination change compensation in relation to encoding and decoding of a multiview image is illustrated.

Referring to FIG. 3A, the apparatus 300 for encoding an image according to the present invention includes a motion estimator 302, a motion compensator 308, a light change difference value calculator 310, and a light change difference value predictor 312. , DCT and quantization unit 314, entropy coding unit 316, inverse quantization and inverse DCT unit 318, deblocking unit 320, and frame memory 322. Here, the deblocking unit 320 corresponds to the deblocking filtering device according to the present invention.

The lighting change difference value calculating unit 310 illuminates by dividing an average value of a current macroblock including a current block to be encoded and a mean value of a macroblock including a reference block to be used for inter prediction of the current block. Calculate the difference value of illumination change (DVIC). Here, the macro block is an example of an image processing unit, and hereinafter, it is assumed that the macro block is an image processing unit. In the lighting change compensation, the lighting change difference value is generated in an image processing unit, that is, macroblock units.

The average value of the macro block including the current macro block and the reference block is calculated, and the difference between the two calculated values is calculated to calculate the difference in lighting change. The illumination change difference value prediction unit 312 subtracts the predicted illumination change difference value predicted from the neighboring blocks of the current macro block from the calculated illumination change difference value and inserts it into the bitstream to predictively encode the difference.

The motion estimation unit 302 estimates the motion vector of the current block based on the lighting change difference value calculated by the lighting change difference value calculating unit 310. The illumination change compensator 304 compensates for the illumination change based on the difference in the illumination change, and the motion vector determiner 306 uses the current block compensated for the illumination change to the frame memory 322. The motion vector is estimated by searching the stored reference picture. The estimated motion vector is inserted into the bitstream and transmitted to the image decoding apparatus.

The motion compensator 308 performs motion compensation using the average value of the current macro block, the average value of the macro block including the reference block, and the motion vector to generate a prediction block that is a prediction value of the current block. Motion compensation may be performed through Equation 1.

은 현재 매크로 블록의 화소 평균값,

는 참조 블록이 포함되어 있는 매크로 블록의 화소 평균값이며, (x', y')은 움직임 벡터이다.NewR (i, j) means the residue at (i, j) coordinates generated using illumination change compensation, f (i, j) is the pixel value at (i, j) coordinates of the current block, r (i + x ', j + y') is a pixel value in the (i + x ', j + y') coordinate of the reference picture,

Is the average pixel value of the current macro block,

Is the pixel average value of the macro block including the reference block, and (x ', y') is a motion vector.

The generated residue is quantized after the discrete cosine transformed by the DCT and quantization unit 314 and orthogonally transformed into the frequency domain. The quantized residue is variable length coded by the entropy encoding unit 316 and inserted into the bitstream.

The inverse quantization unit and the DCT unit 318 inversely quantize the quantized residue and then inverse discrete cosine transform to restore the residue before encoding. The reconstructed residue is added to the prediction block of the current block generated by the motion compensator 308 and reconstructed into a block before encoding.

The reconstructed block is stored in the frame memory 322 to be used as a reference picture, which is stored after the block boundary is deblocked and filtered by the deblocking unit 320 to remove blocking artifacts. Here, if the deblocking filtering according to the prior art is performed as it is, the blocking artifacts are removed since the current image processing unit and / or another image processing unit adjacent to the current image processing unit are not considered to be encoded using the illumination change compensation. It can happen if it doesn't. Therefore, a deblocking filtering method is required to remove this and a detailed description thereof will be described later with reference to FIGS. 4 to 6.

3B illustrates an apparatus for decoding an image according to an embodiment of the present invention. An apparatus for decoding an encoded image by performing inter prediction using lighting change compensation is illustrated.

Referring to FIG. 3B, the apparatus 350 for decoding an image according to the present invention includes an entropy decoding unit 352, an inverse quantization and inverse DCT unit 354, an illumination change difference value predictor 356, and a motion compensator 358. And a deblocking unit 360 and a frame memory 362. Here, the deblocking unit 360 corresponds to the deblocking filtering device according to the present invention.

The entropy decoding unit 352 entropy decodes the data of the residue of the current block included in the bitstream. The data for the entropy decoded residue is inverse quantized by the inverse quantization and inverse DCT unit 354, inverse discrete cosine transform, and restored. The reconstructed residue is added to the predicted block generated by the motion compensator 358 to restore the current block.

The motion compensator 358 corresponds to the motion compensator 308 of the encoding apparatus. The motion compensator 358 uses the illumination change difference value in generating the prediction block of the current block. The lighting change difference value predicted from the neighboring blocks of the current macro block including the current block is added to the predicted coded lighting change difference value included in the bitstream by adding the difference of the current block. Generates the difference in illumination variation.

In addition, a reference picture stored in the frame memory 362 is searched using a motion vector included in the bitstream to specify a macroblock including the reference block, and a difference in illumination change generated at an average value of the specified macroblock is obtained. By adding the values, we generate the average value of the current macro block. A prediction block of the current block is generated using the average value of the generated current macroblock and the reference block.

The current block reconstructed by adding the predicted block and the residue generated by the motion compensator 358 is stored in the frame memory 362 to be used as a reference picture. The deblocking unit 360 removes the blocking artifacts. The deblocking is stored after filtering.

4 to 6 are flowcharts illustrating a deblocking filtering method according to an embodiment of the present invention. 3A and 3B are flowcharts illustrating a method of performing deblocking filtering by the deblocking unit 320 or 360.

4 is a flowchart illustrating a deblocking filtering method according to an embodiment of the present invention. This will be described in detail with reference to FIG. 7.

Referring to FIG. 4, in operation 410, the deblocking unit 320 or 360 according to the present invention is adjacent to a difference in illumination variation of the current image processing unit 710 including the current block 711 and a current image processing unit. It is determined whether an illumination change difference value of another image processing unit 720 or 730 is different. Here, the current image processing unit is preferably a macro block.

If the current image processing unit 710 and / or the adjacent image processing unit 720 or 730 are encoded using the illumination change compensation, blocking artifacts are generated. Perform blocking filtering.

Therefore, in operation 410, it is determined whether the lighting change difference value of the current image processing unit 710 is different from the lighting change difference value of another image processing unit 720 or 730 adjacent to the current image processing unit.

If it is determined whether the lighting change difference value of the current block 711 and the lighting change difference values of the neighboring blocks 712 to 715 adjacent to the current block are different from each other, the current block is present as shown in FIG. 7. In the case of the block located inside the image processing unit 710, the difference value of the illumination change is the same as the neighboring blocks 712 to 715. The lighting change difference value is calculated in the image processing unit, that is, macroblock unit, and since the deblocking filtering is performed in the subblock unit of the macroblock, the lighting change difference value of the current block 711 and the neighboring blocks 712 to 715 is different. Is the same. Therefore, the illumination change difference values are compared in the image processing units 710 to 730.

Preferably, the vertical boundary of the current block, that is, the boundary between the current block 711 and the block 713 adjacent to the left side of the current block and the boundary between the current block 711 and the block 715 adjacent the right side of the tongue block. When the deblocking filtering is performed, it is determined whether an illumination change difference value of the current image processing unit 710 and an illumination change difference value of the image processing unit 720 adjacent to the left side of the current image processing unit 710 are different.

In addition, the horizontal boundary of the current block, that is, the boundary between the current block 711 and the block 712 adjacent to the top of the current block and the boundary between the current block 711 and the block 714 adjacent to the bottom of the current block. When the deblocking filtering is performed, it is determined whether the difference in illumination change of the current image processing unit 710 and the difference in illumination change of the image processing unit 730 adjacent to the upper portion of the current image processing unit 710 are different.

If it is determined in step 410 that the difference in illumination change of the current image processing unit 710 and the difference in illumination change of the adjacent image processing unit 720 or 730 are different, in step 420 the filtering intensity is set to Bs = K ₁ to determine the present value. Deblock the filtering of the block boundaries.

On the contrary, if it is determined in step 410 that the difference in illumination change of the current image processing unit 710 and the difference in illumination change of the adjacent image processing unit 720 or 730 are not different, the filtering intensity is set to Bs = K ₂ in step 430. Set to deblock and filter the boundary of the current block.

Based on the determination result in step 410, K ₁ and K ₂ are set to different values to perform deblocking filtering with different filtering strengths. In addition, when the illumination change difference values are different, it is preferable to perform the deblocking filtering by making the filtering intensity harder. In other words, set K ₁ to a value greater than K ₂ .

There may be a case where one of the current image processing unit 710 and one of the adjacent image processing units 720 and 730 is not encoded using the lighting change compensation. The blocking artifacts according to the lighting change compensation may occur even when the image processing unit encoded using the lighting change compensation and the image processing unit encoded without using the lighting change compensation are adjacent to each other. Therefore, in order to perform the deblocking filtering in consideration of such a case, the deblocking unit 320 or 360 according to the present invention sets the difference in the illumination change of the encoded image processing unit to '0' without using the illumination change compensation. The determination of step 410 is performed.

If both the current image processing unit and the adjacent image processing unit are encoded without using the illumination change compensation, since the illumination change difference values are all '0', it is determined that the illumination change difference values are the same and the filtering intensity is determined in step 430 by Bs = K. Set to ₂ to deblock and filter the boundaries of the current block.

However, when one of the current image processing unit and the adjacent image processing unit is encoded using the lighting change compensation, the lighting change difference value of the image processing unit encoded using the lighting change compensation is not '0', so the lighting change difference In operation 420, the filtering strength is set to Bs = K ₁ , and the deblocking filtering is performed on the boundary of the current block.

5 is a flowchart illustrating a deblocking filtering method according to another embodiment of the present invention.

Step 510 is the same step as step 410 of FIG. 4, where the deblocking unit 320 or 360 is different from the illumination change difference value of the current image processing unit 710 including the current block 711 and other adjacent to the current image processing unit. It is determined whether an illumination change difference value of the image processing unit 720 or 730 is different.

If it is determined in step 510 that the difference in the illumination change of the current image processing unit 710 and the difference in the illumination change in the adjacent image processing unit 720 or 730 are different, the deblocking unit 320 or 360 determines in step 520 the current image. The absolute value of the difference between the illumination change difference value of the processing unit 710 and the illumination change difference value of the adjacent image processing unit 720 or 730 is compared with the first threshold value T ₁ .

As a result of the comparison, if the absolute value is less than the first threshold value T ₁ , the absolute value is again compared with the second threshold value T ₂ in step 530. Here, the second threshold T ₂ is smaller than the first threshold T ₁ . There is no limit to the number of thresholds for comparing the absolute values, and an embodiment may be configured such that only the first threshold value T ₁ is compared and not the second threshold value T ₂ . If the first comparison threshold (T ₁₎ and only there, if the absolute value is smaller than the first threshold (T _1), by setting the filter strength in Step 560 to the bar as Bs = K ₃ performs a deblocking filtering .

If the absolute value of the comparison in step 530 is greater than the second threshold value T ₂ , in other words, if the absolute value has a value between the first threshold value T ₁ and the second threshold value T ₂ , in step 550 Deblocking filtering is performed by setting the filtering strength to Bs = K ₂ .

In the case of steps 540 and 550, deblocking filtering may be performed by setting K ₁ and K ₂ to different values to have different filtering strengths. However, in the case of steps 560 and 570, the embodiment may be configured to adjust the filtering strength again according to the deblocking filtering method according to the prior art without assigning separate filtering strength values to K ₃ and K ₄ . Detailed description thereof will be described later with reference to FIG. 6.

6 is a flowchart illustrating a deblocking filtering method according to another embodiment of the present invention. 6 illustrates a deblocking filtering method in which the deblocking filtering method according to the present invention is applied to the deblocking filtering method according to the prior art.

Referring to FIG. 6, steps 603 to 605 are steps according to the deblocking filtering method of the present invention. Assume that the current block 711 is a p block, and another block 712 or 713 or 714 or 715 adjacent to the current block is a q block. A method of deblocking filtering of the boundary of the current block 711, that is, the boundary between the current block and another block 712 or 713 or 714 or 715 adjacent to the current block will be described.

Step 601 is the same as step 201 shown in FIG. 2, and the deblocking unit 320 or 360 according to the present invention determines whether at least one of two adjacent blocks p and q is intra coded.

If it is determined in step 601 that at least one of the adjacent blocks p and q is an intra-coded block, it is determined again in step 602 whether the boundary of blocks p and q is a macro block boundary, and in step 609 or In operation 610, deblocking filtering is performed according to different filtering strengths.

If it is determined in step 601 that the adjacent blocks p and q are both non-intra coded blocks, in step 603 the difference between the lighting changes and the current value of the current image processing unit 710 including the p block, that is, the current block 711, is present. It is determined whether the illumination change difference values of the other image processing units 720 and 730 adjacent to the image processing unit are different.

If it is determined in step 603 that the difference in illumination change of the current image processing unit 710 and the difference in illumination change of another image processing unit 720 or 730 adjacent to the current image processing unit are not different, deblocking according to the prior art. Like the filtering method, the strength of the deblocking filtering is adjusted through steps 606 to 608.

If it is determined in step 603 that the illumination change difference value of the current image processing unit 710 and the illumination change difference value of another image processing unit 720 or 730 adjacent to the current image processing unit are not different from each other, the illumination is performed in steps 604 and 605. The absolute value for the difference of the change difference values is compared with predetermined thresholds.

If it is determined that the absolute value is greater than the first threshold value T ₁ , the deblocking filtering is performed by setting the filtering strength to Bs = K ₁ in step 611. Comparison result if the absolute value is determined as a value between the first threshold (T ₁₎ and the first threshold value is smaller second threshold value that is less than (T ₁₎ (T _2), in step 612 the filter strength Bs = K ₂ Set to perform deblocking filtering.

In addition, if it is determined that the absolute value is smaller than the first threshold value T ₁ and the second threshold value T ₂ , the filtering strength is adjusted again according to the deblocking filtering method according to the prior art. In other words, steps 203 to 205 shown in FIG. 2 are performed the same in steps 606 to 608.

The combination of the deblocking filtering method according to the present invention and the deblocking filtering method according to the prior art shown in FIG. 6 are just examples, and various other combinations are possible, and can be easily understood by those skilled in the art. Can be.

8 illustrates syntax of a slice header according to an embodiment of the present invention.

Referring to FIG. 8, a slice header according to an embodiment of the present invention includes syntax for a first threshold value T ₁ and / or a second threshold value T ₂ . The encoding side includes information about the first threshold value T ₁ and / or the second threshold value T ₂ in the slice header. 'slice_ic_th1_mvc' is information about the first threshold value T ₁ , and 'slice_ic_th2_mvc' is information about the second threshold value T ₂ . The decoding side sets the first threshold value T ₁ and the second threshold value T ₂ with reference to the slice header, and deblocks the boundary of the current block by applying the deblocking filtering method illustrated in FIGS. 4 to 6. To filter.

When the slice header does not include the information about the first threshold value T ₁ and / or the second threshold value T ₂ , deblocking filtering is performed by using a threshold value determined as a default. It can also be configured. For example, if there is no information on the first threshold (T ₁₎ in the slice header, the first predetermined threshold (T ₁₎ to "3", and the second it does not have information about the threshold value (T ₂₎ In this case, the deblocking filtering may be performed by setting the second threshold value T ₂ to '1'.

9 illustrates a deblocking unit 320 or 360 according to an embodiment of the present invention.

Referring to FIG. 9, the deblocking unit 320 or 360 according to the present invention includes a control unit 910 and a filtering unit 920.

The controller 910 may be configured to display a difference in illumination change of the current image processing unit 710 including the current block 711 and a difference in illumination change of another image processing unit 720 or 730 adjacent to the current image processing unit 710. Determine if this is different. If judged to be different, the absolute value for the difference of the lighting change difference values is compared with a predetermined threshold (s).

If the current image processing unit 710 and / or the other image processing unit 720 or 730 are not encoded using the lighting change compensation, set the lighting change difference value to '0' and determine whether the lighting change difference value is different. .

The filtering unit 920 performs deblocking filtering by adjusting the filtering strength based on the determination or comparison result of the control unit 710.

Deblocking the boundary of the current block 711 by applying different filtering intensities depending on whether the difference in illumination variation of the current image processing unit 710 and the difference in illumination variation of another image processing unit 720 or 730 are different. To filter.

In addition, deblocking filtering is performed by comparing the absolute value of the difference between the illumination change difference values with a predetermined threshold value (s) and applying different filtering intensities based on the comparison result.

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, the spirit of the invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications will fall within the scope of the invention. In addition, the system according to the present invention can be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of the 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 include a carrier wave (for example, transmission through the Internet). The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

According to the present invention, since the deblocking filtering may be performed in consideration of the case where the current image processing unit and / or another image processing unit adjacent to the current image processing unit are encoded using the lighting change compensation, a new generation is generated by the lighting change compensation. You can eliminate blocking artifacts.

In addition, the deblocking filtering may be performed in consideration of whether the subblocks located at the boundary of the image processing unit are encoded using the illumination change compensation on the subblocks located inside the image processing.

Claims (19)

In the deblocking filtering method of the image, Determining whether a difference value of illumination change of a current image processing unit including a current block is different from a difference value of illumination change of another image processing unit adjacent to the current image processing unit; And And deblocking filtering the boundary of the current block by adjusting a filtering strength of the deblocking filter based on the determination result. The method of claim 1, wherein the image processing unit Deblocking filtering method characterized in that the macro block. The method of claim 1, wherein the determining And comparing an absolute value of a difference between an illumination change difference value of the current image processing unit and an illumination change difference value of the other image processing unit than a first threshold value. The method of claim 3, wherein the filtering step And deblocking filtering by applying different filtering intensities when the absolute value is larger and smaller than the first threshold value. 4. The method of claim 3, wherein the comparing step And if the absolute value is less than the first threshold value, comparing the absolute value with a second threshold value that is less than the first threshold value. The method of claim 5, wherein the filtering step And deblocking filtering by applying different filtering intensities when the absolute value is larger and smaller than the second threshold value. The method of claim 1, wherein the determining Setting an illumination change difference value of an encoded image processing unit to '0' without using illumination compensation among the current image processing unit and the other image processing unit; And And determining whether an illumination change difference value of the current image processing unit and an illumination change difference value of the other image processing unit are different based on the result of the setting. The method of claim 1, The determining step Determining whether a difference value of illumination change of the current image processing unit is different from a difference value of illumination change of another image processing unit adjacent to a left side of the current image processing unit, The filtering step And deblocking filtering the vertical boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result. The method of claim 1, The determining step Determining whether a difference value of illumination change of the current image processing unit is different from a difference value of illumination change of another image processing unit adjacent to an upper portion of the current image processing unit; The filtering step And deblocking filtering the horizontal boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result. In the deblocking filtering apparatus of the image, A controller for determining whether a difference value of illumination change of the current image processing unit including the current block is different from a difference in illumination change of another image processing unit adjacent to the current image processing unit; And And a filtering unit which deblocks and filters the boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result. The method of claim 10, wherein the image processing unit Deblocking filtering device, characterized in that the macro block. The method of claim 10, wherein the control unit And comparing an absolute value of a difference between an illumination change difference value of the current image processing unit and an illumination change difference value of the other image processing unit from a first threshold value. The method of claim 12, wherein the filtering unit And deblocking filtering by applying different filtering intensities when the absolute value is larger and smaller than the first threshold value. The method of claim 12, wherein the control unit And when the absolute value is less than the first threshold value, comparing the absolute value with a second threshold value less than the first threshold value. The method of claim 14, wherein the filtering unit And when the absolute value is greater than or less than the second threshold, deblocking filtering by applying different filtering intensities. The method of claim 10, wherein the control unit Among the current image processing unit and the other image processing unit, the illumination change difference value of the image processing unit is set to '0' without using illumination compensation, and based on the setting result, the current image processing unit And determining whether an illumination change difference value between the illumination change difference value and the illumination change difference value of the other image processing unit is different. The method of claim 10, The control unit Determining whether a difference value of illumination change of the current image processing unit is different from a difference value of illumination change of another image processing unit adjacent to the left side of the current image processing unit, The filtering unit And deblocking filtering the vertical boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result. The method of claim 10, The control unit Determining whether a difference value of illumination change of the current image processing unit is different from a difference value of illumination change of another image processing unit adjacent to an upper portion of the current image processing unit, The filtering step And deblocking filtering the horizontal boundary of the current block by adjusting the filtering strength of the deblocking filter based on the determination result. A computer readable recording medium having recorded thereon a program for executing the method of any one of claims 1 to 9.

Images