KR20100046732A - Video encoding/decoding apparatus, deblocking filter and deblocing filtering method, and recording medium therefor - Google Patents

Abstract

The present invention relates to a video encoding / decoding apparatus, a deblocking filtering apparatus and method for the same, and a recording medium. The deblocking filtering apparatus according to an embodiment of the present invention determines a strength of a boundary between subblocks of a current block. A boundary strength determiner; An optimal direction may be selected based on the amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block, and the predetermined direction may be selected for all pixels of the boundary of the sub-block. A filtering direction determiner configured to determine, as a filtering direction of the corresponding sub-block, the most selected direction among a plurality of optimal directions; And selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and determining whether to filter the determined filtering direction according to the relationship of pixel values between the selected pixels. And a filtering unit configured to perform filtering on the selected pixels in the determined filtering direction according to whether to filter or not, thereby improving block performance and removing computation.

Description

Video encoding / decoding apparatus, apparatus and method for deblocking filtering, and recording medium for video encoding / decoding apparatus, and deblocking filtering method and recording medium therefor}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video encoding / decoding apparatus, and more particularly, to a video encoding / decoding apparatus for removing a blocking artifact occurring in a device encoding using block-based discrete cosine transform and quantization. A blocking filtering device, a filtering method, and a recording medium.

In general, in most commercial video compression schemes and devices (H.263, H.264, MPEG-4, H.264 / AVC), block-based discrete cosine transforms are used to reduce spatial redundancy of images. DCT) and quantization techniques. In this case, the transforms of block units that do not overlap each other are independently subjected to discrete cosine transform and quantization without any correlation between neighboring blocks or pixels, thereby causing data loss, and in particular, greatly increasing the quantization value. In low bit-rate video, the blocking phenomenon, which is a structured discontinuity of pixel values, appears clearly at the interface between blocks, which is called blocking artifact. Blocking artifacts not only significantly reduce the picture quality of the video screen, but are also stored in the frame memory with block distortion, and the image with block distortion is referenced in the motion compensation process, so that image quality deterioration is propagated, which greatly reduces the compression performance. There is a problem.

For this reason, this problem can be solved by using a low-band filter that reduces the blocking artifacts before the decoder stores the decoded image in the reference frame memory and also improves the image quality in the decoder. The technique is called deblocking filter.

However, if there is an edge present in the image or an object boundary in the screen at the block boundary surface, applying the deblocking filter to this may cause the edge portion of the original image to be damaged, thereby causing deterioration of image quality.

Therefore, the deblocking filter in the H.264 / AVC video codec adaptively performs filtering for each boundary of a block having 4x4 pixels, which is a block unit performing quantization with DCT. In more detail, if the boundary of the block is determined to be the edge of the real image or the boundary of the object, the filtering is not performed. If the boundary is determined to be distortion due to blocking artifacts, the problem is solved.

However, since the H.264 / AVC deblocking filter determines the existence of an edge only in the vertical and horizontal directions, when a boundary of an edge or an object is placed in a diagonal direction, discontinuity occurs in diagonal pixels of the boundary between blocks. There is still a problem that blocking artifacts can occur.

In order to solve the above problems, the present invention, by determining the edge direction in the pixel unit in the vertical and horizontal direction as well as a number of other directions at the boundary between blocks, and remove the blocking artifacts by performing deblocking filtering in that direction A video encoding / decoding device, a deblocking filtering device and a filtering method therefor, and a recording medium, which improve image quality, and in particular, reduce the amount of computation occurring when removing blocking artifacts, thereby facilitating implementation in a limited amount of computation. For the purpose of

According to an aspect of an embodiment of the present invention to achieve the above object, a boundary strength determination unit for determining the strength of the boundary between the sub-blocks of the current block; An optimal direction may be selected based on the amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block, and the predetermined direction may be selected for all pixels of the boundary of the sub-block. A filtering direction determiner configured to determine, as a filtering direction of the corresponding subblock, the most selected direction among a plurality of optimal directions; And selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and determining whether to filter the determined filtering direction according to the relationship of pixel values between the selected pixels. A deblocking filtering device including a filtering unit which performs filtering on the selected pixels in the determined filtering direction according to filtering is provided.

The filtering direction determiner may include an absolute value of a pixel value difference between first pixels facing the boundary, an absolute value of a pixel value difference between the second pixel and the first pixel on the basis of the boundary, An absolute value of a pixel value difference between the second pixel and the first pixel on the other side based on the corresponding boundary, an absolute value of a pixel value difference between the third pixel and the second pixel on the side based on the corresponding boundary, and the corresponding value The change amount may be calculated based on an absolute value of a difference in pixel values between the third pixel and the second pixel on the other side of the boundary, and a direction in which the change amount is the smallest may be selected as the optimum direction.

The filtering unit may include an absolute value of a difference of pixel values between first pixels facing a corresponding boundary, and a relationship between a second pixel and a first pixel of one side based on the boundary. If the absolute value of the pixel value difference and the absolute value of the pixel value difference between the second pixel and the first pixel on the other side are smaller than predetermined thresholds, the filter is performed in the determined filtering direction. Can be.

If the determined filtering direction does not satisfy the condition, the filtering unit may determine whether the condition is satisfied with respect to the vertical direction of the corresponding boundary, and may perform filtering according to the check result.

The preset plurality of directions may include 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees, and 135 degrees.

The filtering direction determiner may record the determined filtering direction as additional information about the corresponding subblock, or allocate additional information about pixels filtered in the determined filtering direction among pixels on the boundary of the corresponding subblock.

According to another aspect of an embodiment of the present invention to achieve the above object, a boundary strength determination unit for determining the strength of the boundary between the sub-blocks of the current block; For the first pixel of the boundary of the sub-block, one direction is determined as the filtering direction based on the amount of change in the pixel value between the pixels corresponding to each direction among a plurality of preset directions, and two of the boundaries of the sub-block are determined. For the second or more pixels, the filtering direction is determined as the filtering direction for the current subblock, the filtering direction of the immediately preceding pixel is determined as the filtering direction, or the filtering direction is determined according to the filtering direction determination process of the first pixel. A filtering direction determiner; And selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and determining whether to filter the determined filtering direction according to the relationship of pixel values between the selected pixels. A deblocking filtering device including a filtering unit that performs filtering on the selected pixels in the determined filtering direction according to filtering is provided.

According to another aspect of an embodiment of the present invention to achieve the above object, the boundary strength determination step of determining the strength of the boundary between sub-blocks of the current block; An optimal direction selecting step of selecting an optimum direction based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block; A filtering direction determining step of determining, as the filtering direction of the corresponding sub block, the most selected direction among the selected plurality of optimal directions for all pixels on the boundary of the sub block; A pixel selecting step of selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering step determining step of determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering step of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering.

According to another aspect of an embodiment of the present invention to achieve the above object, the boundary strength determination step of determining the strength of the boundary between sub-blocks of the current block; A first filtering direction determining step of determining, with respect to the first pixel of the boundary of the sub block, one direction as a filtering direction based on an amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions; For the second or more pixels on the boundary of the sub-block, the direction most selected so far for the current sub-block is determined as the filtering direction, or the filtering direction of the immediately preceding pixel is determined as the optimal direction, or the filtering direction of the first pixel is determined. Determining a filtering direction according to the determining process; A pixel selecting step of selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering step determining step of determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering step of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering.

According to another aspect of an embodiment of the present invention to achieve the above object, a prediction unit for predicting the current block of the image to generate a prediction block; A subtraction unit for generating a residual block by subtracting the prediction block from the current block; A transformer for converting the residual block into a frequency domain; A quantizer for quantizing the transformed residual block; An encoder which encodes the quantized residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block into a residual block having a pixel signal on a time axis; An adder configured to add the prediction block to the inverse transformed residual block to restore a current block; And deblocking filtering the reconstructed current block, based on an amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of a boundary of a subblock of the reconstructed current block. Is selected, and the most selected direction among the selected plurality of optimal directions for all pixels of the boundary of the sub block is determined as the filtering direction of the corresponding sub block, and the determined filtering is performed on each pixel of the boundary of the sub block. Selecting pixels to filter based on a direction, determining whether to filter the determined filtering direction according to the relationship between the pixel values between the selected pixels, and determining whether to filter the determined filtering direction according to the determined filtering. Provided is a video encoding apparatus including a deblocking filter that performs filtering on a do.

According to another aspect of an embodiment of the present invention to achieve the above object, a prediction unit for predicting the current block of the image to generate a prediction block; A subtraction unit for generating a residual block by subtracting the prediction block from the current block; A transformer for converting the residual block into a frequency domain; A quantizer for quantizing the transformed residual block; An encoder which encodes the quantized residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block into a residual block having a pixel signal on a time axis; An adder configured to add the prediction block to the inverse transformed residual block to restore a current block; And deblocking filtering the reconstructed current block, and for the first pixel of the boundary of the sub-block of the reconstructed current block, based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions. One direction is determined as the filtering direction, and for the second or more pixels on the boundary of the sub block, the direction most selected so far is determined as the filtering direction for the current sub block, or the filtering direction of the immediately preceding pixel is determined as the filtering direction. Alternatively, the filtering direction is determined according to the filtering direction determination process of the first pixel, and pixels to be filtered are selected for each pixel of the boundary of the sub-block based on the determined filtering direction, and pixel values between the selected pixels are selected. Determine whether to filter the determined filtering direction according to the relationship between The video encoding apparatus includes a deblocking filter to perform filtering on the selected pixel of the determined filtering direction is provided, depending on whether the determined filter.

According to another aspect of an embodiment of the present invention to achieve the above object, a decoding unit for decoding the bitstream to extract the residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And deblocking filtering the restored current block, and for pixels corresponding to the received additional information among all the pixels on the boundary of the subblock of the restored current block, between pixels corresponding to each of a plurality of preset directions. Selecting an optimal direction based on the amount of change in the pixel value, determining the selected optimal direction as a filtering direction of the corresponding subblock, and filtering pixels based on the determined filtering direction for each pixel at the boundary of the subblock. Deblocking for filtering the selected filtering direction according to the relationship between the pixel values of the selected pixels, and performing filtering on the selected pixels in the determined filtering direction according to the determined filtering. Provided is a video decoding apparatus including a filter unit.

According to another aspect of an embodiment of the present invention to achieve the above object, a decoding unit for decoding the bitstream to extract the residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And deblocking and filtering the current block reconstructed by the adder, and for each pixel of the boundary of the subblock of the reconstructed current block, the pixels to be filtered based on the filtering direction received as additional information of the corresponding subblock. Selects and determines whether to filter the received filtering direction according to the relationship between the pixel values between the selected pixels, and performs filtering on the selected pixels of the received filtering direction according to the determined filtering. A video decoding apparatus including a blocking filter unit is provided.

According to another aspect of an embodiment of the present invention to achieve the above object, a decoding unit for decoding the bitstream to extract the residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And deblocking filtering the current block reconstructed by the adder, wherein the first pixel of the boundary of the subblock of the reconstructed current block has a pixel value between pixels corresponding to each direction among a plurality of preset directions. Based on the amount of change, one direction is determined as the filtering direction, and for the second or more pixels on the boundary of the sub block, the direction most selected so far is determined as the filtering direction for the current sub block, or the filtering direction of the immediately preceding pixel is determined. Determine a filtering direction according to a filtering direction or a filtering direction determination process of the first pixel, and select pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and the selected pixel Filtering for the determined filtering direction according to the relationship between pixel values A crystal, and the moving image decoding apparatus that includes the determined for the selected pixel of the determined direction filtering deblocking filter to perform filtering based on whether the filtering portion is provided.

According to another aspect of an embodiment of the present invention to achieve the above object, a boundary strength determination function for determining the strength of the boundary between sub-blocks of the current block; An optimum direction selection function for selecting an optimum direction based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block; A filtering direction determining function for determining, as the filtering direction of the corresponding sub block, the most selected direction among the plurality of selected optimal directions for all pixels on the boundary of the sub block; A pixel selection function for selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering or not determining function for determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering function for performing filtering on the selected pixels in the determined filtering direction in accordance with the determined filtering.

According to another aspect of an embodiment of the present invention to achieve the above object, a boundary strength determination function for determining the strength of the boundary between sub-blocks of the current block; A first filtering direction determining function for determining a first direction as a filtering direction with respect to a first pixel of a boundary of the sub block based on a change amount of a pixel value between pixels corresponding to each direction among a plurality of preset directions; For the second or more pixels on the boundary of the sub-block, the direction most selected so far for the current sub-block is determined as the filtering direction, or the filtering direction of the immediately preceding pixel is determined as the optimal direction, or the filtering direction of the first pixel is determined. A second filtering direction determining function for determining a filtering direction according to the determining process; A pixel selection function for selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering or not determining function for determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering function for performing filtering on the selected pixels in the determined filtering direction in accordance with the determined filtering.

According to an embodiment of the present invention, a technique for applying a deblocking filter on a pixel-by-pixel basis for not only a vertical horizontal direction but also a plurality of other directions is provided, so that a diagonal discontinuity caused by an edge with respect to a plurality of diagonal directions is provided. It effectively removes the block distortion caused by sex, and in particular, effectively reduces the amount of computation that occurs when removing the block distortion.

Hereinafter, some embodiments of the present invention will be described in detail through exemplary drawings. In adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are assigned to the same components as much as possible even though they are shown in different drawings. In addition, in describing the present invention, when it is determined that the detailed description of the related well-known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, in describing the component of this invention, terms, such as 1st, 2nd, A, B, (a), (b), can be used. These terms are only for distinguishing the components from other components, and the nature, order or order of the components are not limited by the terms. If a component is described as being "connected", "coupled" or "connected" to another component, that component may be directly connected to or connected to that other component, but there may be another configuration between each component. It is to be understood that the elements may be "connected", "coupled" or "connected".

Since the video screen is composed of 30 frames in one second, the difference is small between one frame and neighboring frames, and thus cannot be distinguished by the human eye. For this reason, when 30 frames are sprayed in one second, the human eye recognizes the frames as continuous.

As such, when the previous frame is similar to the current frame, the pixel value of the next frame can be predicted from the known pixel values constituting the previous frame. This is called inter-prediction.

The encoding and decoding of such video data is performed based on a motion prediction technique. The motion prediction is performed by referring to past frames or referring to both past and future frames based on the time axis. The frame referred to to encode or decode the current frame is called a reference frame. In block-based video encoding, one still image (frame) constituting a video is divided into macroblocks and subblocks constituting the macroblock, and motion is predicted in units of blocks and encoding is performed.

It is also possible to predict the next pixel using the correlation of the pixel signal within the same frame and to encode the prediction error. This is called intra-prediction.

1 is a block diagram of a video encoding apparatus according to an embodiment of the present invention.

The video encoding apparatus 100 according to an embodiment of the present invention includes a predictor 110, a subtractor 120, a transformer 130, a quantizer 140, an encoder 150, and an inverse quantizer 160. ), An inverse transform unit 170, an adder 180, and a deblocking filter unit 190.

The video encoding apparatus 100 may be a personal computer (PC), a notebook computer, a personal digital assistant (PDA), a portable multimedia player (PMP), or a PlayStation Portable (PSP). ), A communication device such as a communication modem for communicating with various devices or a wired / wireless communication network, a memory for storing various programs and data for encoding an image, and executing a program. Means a variety of devices including a microprocessor for operation and control.

As described above, the prediction unit 110 may use the current prediction using any one or a combination of intra prediction based on motion prediction or intra prediction that predicts the next pixel using the correlation of pixel signals within the same frame. A block (or macroblock) can be predicted.

For example, the predictor 110 may be divided into a motion estimator (not shown) and a motion compensator (not shown). The motion estimator finds the motion prediction value of the macroblock of the current frame in the reference frame and outputs the difference of the motion as a motion vector. That is, the macroblock to be searched is searched within a predetermined search area of the reference frame to find the most similar macroblock and the degree of movement thereof is output as a motion vector. The motion compensation unit obtains a prediction macro block corresponding to the obtained motion vector from the reference frame.

As another example, the predictor 110 is an intra predictor that predicts a current macroblock of a current frame using a neighboring macroblock of a current macroblock in an image, and at least one pixel value of at least one neighboring macroblock. The predicted macroblock is predicted by calculating a predicted pixel value of each pixel of the current macroblock using. Here, the peripheral macroblock may be one or more peripheral macroblocks that are compressed before the current macroblock and located around the current macroblock.

The subtractor 120 subtracts the prediction block from the current block to generate a residual block. That is, the subtractor 120 generates a residual block having a residual signal by calculating a difference value between an original pixel value of each pixel of the current block and a predicted pixel value of each pixel of the prediction block. do.

The transformer 130 transforms the residual block into the frequency domain. That is, the transformer 130 converts the residual block into a frequency domain to generate a residual block having a frequency coefficient. Here, the transform unit 130 transforms the residual block, and converts various image signals of a time axis into a frequency axis such as a Hadamard transform, a discrete cosine transform based transform, or the like. The residual signal may be converted into the frequency domain using a conversion technique, and the residual signal converted into the frequency domain becomes a frequency coefficient.

The quantizer 140 quantizes the residual block transformed by the transformer 130. That is, the quantization unit 140 quantizes the frequency coefficients of the residual block to generate quantization frequency coefficients. Here, the quantization unit 140 uses Dead Zone Uniform Threshold Quantization (DZUTQ), Quantization Weighted Matrix, or the quantization method to quantize the frequency coefficient of the residual block. Etc. can be used.

The encoder 150 generates a bitstream by encoding the frequency coefficients of the residual block quantized by the quantizer 140. In addition, when the motion vector and the rotation information are transmitted from the predictor 110, the encoder 150 may output the bitstream by encoding the quantized frequency coefficients of the residual block. As the encoding technique, an entropy encoding technique may be used, but various encoding techniques may be used without being limited thereto.

The inverse quantization unit 160 inverse quantizes the residual block quantized by the quantization unit 130. That is, the inverse quantizer 160 inversely quantizes the quantized frequency coefficients of the quantized residual block to generate frequency coefficients.

The inverse transform unit 170 inverse transforms the residual block inversely quantized by the inverse quantization unit 160. That is, the inverse transform unit 170 inversely transforms the frequency coefficients of the inverse quantized residual block to restore the residual block having the pixel signal on the time axis.

The adder 180 reconstructs the current block by adding the prediction block predicted by the predictor 110 to the residual block inversely transformed by the inverse transformer 170.

The deblocking filter 190 deblocks filtering the current block reconstructed by the adder 180. Here, the deblocking filtering refers to an operation of reducing block distortion generated by encoding an image in block units, and applying a deblocking filter to a block boundary and a macroblock boundary, or applying a deblocking filter only to a macroblock boundary or a deblocking filter. You can optionally use one of the methods that does not use. As such, the current block reconstructed by the adder 180 and deblocked filtered by the deblocking filter 190 may be input to the predictor 110 and stored as a reference picture used when predicting the next picture. .

The deblocking filter 190 according to the first embodiment of the present invention deblocks and filters the current block, and in particular, corresponds to each of a plurality of preset directions for each pixel of the boundary of the subblock of the current block. The optimal direction is selected based on the amount of change in the pixel value between the pixels, and the most selected direction among the plurality of optimal directions selected for all the pixels on the boundary of the sub block is determined as the filtering direction of the corresponding sub block, and the sub block is determined. Selecting pixels to be filtered based on the determined filtering direction for each pixel at a boundary of and determining whether to filter the determined filtering direction according to a relationship between pixel values between the selected pixels, and determining whether to filter the determined filtering direction The filtering is performed on the selected pixels in the determined filtering direction. The deblocking filter 190 according to the first embodiment will be described in more detail with reference to FIG. 2.

The deblocking filter 190 according to the second embodiment of the present invention deblocks and filters the current block, and in particular, the first pixel of the boundary of the sub-block of the current block in each of a plurality of preset directions. One direction is determined as the filtering direction based on the amount of change in the pixel value between the corresponding pixels, and the second or more selected pixel as the filtering direction is determined as the filtering direction for the second or more pixels on the boundary of the sub block. Alternatively, the filtering direction of the immediately preceding pixel may be determined as the filtering direction, or the filtering direction may be determined according to the filtering direction determination process of the first pixel, and the filtering direction may be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block. Select pixels and compare the determined filtering direction according to the relationship of pixel values between the selected pixels. And determining whether to filter, and performs filtering on the selected pixel of the determined filtering direction, depending on whether the determined filter. The deblocking filter 190 according to the second embodiment will be described in more detail with reference to FIG. 3.

FIG. 2 is a block diagram of a deblocking filtering device according to a first embodiment of the present invention, and corresponds to the deblocking filter unit 190 according to the first embodiment of FIG.

As shown in FIG. 2, the deblocking filtering device 190 according to the first embodiment of the present invention includes a boundary strength determiner 210, a filtering direction determiner 230, and a filter 250. .

The boundary strength determiner 210 determines the strength of the boundary between sub-blocks of the current block, and determines the boundary strength (BS) according to the characteristics of the block to which pixels belong to the block boundary of the current position. The BS has a value between 0 and 4 and the method of determining is specified in the H.264 / AVC standardization document, for example. That is, the boundary strength determining unit 210 determines the BS value of the boundary surface of each block and determines how much strength to apply filtering according to each BS value. The BS determines whether 0 and 0 depend on the distance between the reference block for motion compensation when the block to which the boundary to be filtered currently belongs uses intra prediction encoding or when using inter prediction encoding. It is determined by a value between four.

As illustrated in FIG. 4, the filtering direction determiner 230 alternates 1, 3, 6, 8, 9, 11, 14, and 16 of 16x16 macroblocks in which 16 4x4 subblocks of 16 to 16 are collected. Deblocking filtering is applied to the first subblock, and a block is applied to each pixel (boundary) (1-16 in Fig. 6) of the boundary of the corresponding subblock (①②③④ in Figs. 5 and 6) for the deblocking filtering. One optimal direction is selected based on the amount of change in the pixel value among the pixels corresponding to each direction among the preset directions (see FIG. 7) based on the boundary, and the subblock The most selected direction among the plurality of selected optimal directions for all the pixels (boundary) of the boundary is determined as the filtering direction of the sub-block. That is, the filtering direction determiner 230 may be, for example, 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees, and the like with respect to a horizontal or vertical boundary as shown in FIG. 7. A direction to filter in nine directions of 135 degrees is preset, and pixels corresponding to nine directions are selected as shown in FIGS. 5 to 13, and the amount of change in pixel values between the pixels corresponding to the respective directions is shown. Then, the direction with the smallest amount of change is selected as the optimum direction, and thus, the optimal direction is selected for each pixel (boundary) of the boundary of the subblock, and then selected for all pixels (boundary) of the boundary of the subblock. The most selected direction among the plurality of optimal directions is determined as the filtering direction of the corresponding subblock. For example, the filtering direction determiner 230 records the filtering direction determined as described above as additional information about the corresponding subblock, or as another example, to one pixel filtered in the filtering direction determined as above among the pixels on the boundary of the corresponding subblock. 1 bit of additional information can be allocated and recorded. This additional information is provided to a decoding apparatus to be described later for use.

For example, referring to FIG. 9, the filtering direction determiner 230 determines an absolute value of the pixel value difference p04-q03 between the first pixels p04 and q03 facing the boundary, and the corresponding boundary. Absolute value of the pixel value difference (p14-p04) between the second pixel p14 and the first pixel p04 on one side as a reference, and the second pixel q13 and the first pixel on the other side based on the corresponding boundary. Absolute value of pixel value difference q13-q03 between (q03), and absolute value of pixel value difference p24-p14 between the third pixel p24 and the second pixel p14 on one side based on the corresponding boundary. The change amount may be calculated based on the absolute value of the pixel value difference q23-q13 between the third pixel q23 and the second pixel q13 on the other side based on the corresponding boundary. A specific example of how to calculate the amount of change in this way will be described later.

The filtering unit 250 selects the pixels to be filtered as shown in FIGS. 8 to 16 based on the filtering direction determined by the filtering direction determiner 230 for each pixel (the boundary) of the boundary of the sub-block. According to the relationship between the pixel values between the pixels to determine whether or not to filter the determined filtering direction, the filtering is performed on the pixels selected as shown in Figures 8 to 16 corresponding to the determined filtering direction according to the determined filtering. .

For example, when the filtering direction determined by the filtering direction determiner 230 is 78.75 degrees of FIG. 9, the filtering unit 250 may have a relationship between pixel values between pixels selected according to the filtering direction of 78.75 degrees of FIG. 9. Absolute value of the pixel value difference (p04-q03) between the first pixels (p04, q03) facing each other, and between the second pixel (p14) and the first pixel (p04) on one side based on the corresponding boundary. The absolute value of the pixel value difference (p14-p04) and the absolute value of the pixel value difference (q13-q03) between the second pixel q13 and the first pixel q03 on the other side are preset based on the boundary. When the condition smaller than the thresholds α and β is satisfied, the filtering is performed in the filtering direction of 78.75 of FIG. 9 determined by the filtering determining unit 230, and the pixel value between the pixels selected according to the filtering direction of 78.75 of FIG. 9 is determined. If the relationship does not satisfy the above conditions, then the Determine to whether or not the condition is satisfied, and it is possible to perform filtering in the vertical direction in accordance with the check result. Specific examples of such filtering or not determination conditions will be described later.

3 is a block diagram of a deblocking filtering device according to a second embodiment of the present invention, and corresponds to the deblocking filter unit 190 according to the second embodiment of FIG.

As shown in FIG. 3, the deblocking filtering device 190 according to the second embodiment of the present invention includes a boundary strength determiner 310, a filtering direction determiner 330, and a filter 350. .

The boundary strength determiner 310 determines the strength of the boundary between sub-blocks of the current block, and determines the boundary strength (BS) according to the characteristics of the block to which pixels belong to the block boundary of the current position. The BS has a value between 0 and 4 and the method of determining is specified in the H.264 / AVC standardization document, for example. That is, the boundary strength determining unit 310 determines the BS value of the boundary surface of each block and determines how much strength to apply filtering according to each BS value. The BS determines whether 0 and 0 depend on the distance between the reference block for motion compensation when the block to which the boundary to be filtered currently belongs uses intra prediction encoding or when using inter prediction encoding. It is determined by a value between four.

As illustrated in FIG. 4, the filtering direction determiner 330 includes 1, 3, 6, 8, 9, 11, 14, and 16 that are alternately arranged in a 16x16 macroblock including 16 16x4x4 subblocks. Deblocking filtering is applied to the subblock, and the first pixel (boundary) of the boundary of the subblock for deblocking filtering (①②③④ of FIGS. 5 and 6) (1 in FIG. 6) is set in advance. Based on the amount of change in the pixel value among the pixels corresponding to each direction among the plurality of directions (see FIG. 7), one direction is determined as the filtering direction, and the second of the boundary of the corresponding sub block is determined. For the above pixels (boundary) (2-16 in FIG. 6), the direction most selected so far for the corresponding subblock is determined as the filtering direction, the filtering direction of the immediately preceding pixel is determined as the filtering direction, or the first pixel Process for determining the filtering direction Depending on the filtering direction can be determined.

For example, referring to FIG. 9, the filtering direction determiner 330 may determine an absolute value of the pixel value difference p04-q03 between the first pixels p04 and q03 facing the boundary, and determine the corresponding boundary. Absolute value of the pixel value difference (p14-p04) between the second pixel p14 and the first pixel p04 on one side as a reference, and the second pixel q13 and the first pixel on the other side based on the corresponding boundary. Absolute value of pixel value difference q13-q03 between (q03), and absolute value of pixel value difference p24-p14 between the third pixel p24 and the second pixel p14 on one side based on the corresponding boundary. And calculating the change amount based on the absolute value of the pixel value difference q23-q13 between the third pixel q23 and the second pixel q13 on the other side with respect to the corresponding boundary, and the calculated amount of change The small direction may be determined as the fill tiling direction. Specific examples of the method of calculating such a change amount will be described later.

The filtering unit 350 selects pixels to be filtered as shown in FIGS. 8 to 16 based on the filtering direction determined by the filtering direction determiner 330 for each pixel (boundary) of the boundary of the corresponding subblock. Determine whether to filter the determined filtering direction according to the relationship between pixel values between the selected pixels, and filter the selected pixels as shown in FIGS. 8 to 16 corresponding to the determined filtering direction according to the determined filtering. To perform.

For example, when the filtering direction determined by the filtering direction determiner 330 is 78.75 degrees in FIG. 9, the filtering unit 350 may have a relationship between pixel values between pixels selected according to the filtering direction of 78.75 degrees in FIG. 9. Absolute value of the pixel value difference (p04-q03) between the first pixels (p04, q03) facing each other, and between the second pixel (p14) and the first pixel (p04) on one side based on the corresponding boundary. The absolute value of the pixel value difference (p14-p04) and the absolute value of the pixel value difference (q13-q03) between the second pixel q13 and the first pixel q03 on the other side are preset based on the boundary. If the condition smaller than the thresholds α and β is satisfied, the filtering is performed in the filtering direction of 78.75 of FIG. 9 determined by the filtering determining unit 230. A specific example of such a filtering or not determination condition will be described later.

17 is a block diagram of a video decoding apparatus according to an embodiment of the present invention.

The video decoding apparatus 1700 according to an embodiment of the present invention includes a decoder 1710, an inverse quantizer 1720, an inverse transform unit 1730, a predictor 1740, an adder 1750, and a deblocking filter unit. 1760.

The video decoding apparatus 1700 may be a personal computer (PC), a notebook computer, a personal digital assistant (PDA), a portable multimedia player (PMP), or a PlayStation Portable (PSP). ), A communication device such as a communication modem for communicating with various devices or a wired / wireless communication network, a memory for storing various programs and data for decoding an image, and executing a program. Means a variety of devices including a microprocessor for operation and control.

The decoder 1710 extracts a residual block by decoding the bitstream. That is, the decoder 1710 decodes a bitstream, which is an image encoded by the image encoding apparatus 100, to extract a residual block, motion vector, and rotation information including pixel information of a current block of the image.

The inverse quantizer 1720 inversely quantizes the residual block extracted by the decoder 1710, and the inverse transformer 1730 inversely transforms the inverse quantized residual block. Here, the inverse quantization and the inverse transform may be inversely transformed with the inverse quantization in the same manner as the inverse quantization and inverse transformation described above with reference to FIG. 1.

The prediction unit 1740 generates a prediction block by predicting the current block by compensating for the motion including the rotation motion of the current block of the image by using the motion vector and the rotation information extracted by the decoder 1710.

The adder 1750 reconstructs the current block by adding the predicted block predicted by the predictor 1740 to the residual block inversely transformed by the inverse transformer 1730.

The deblocking filter unit 1760 deblocks the current block reconstructed by the adder 1750.

The deblocking filter 1760 according to the first embodiment of the present invention deblocks and filters the current block, and in particular, is received from the decoding apparatus 100 as shown in FIG. 1 among all pixels on the boundary of the subblock of the current block. For a specific pixel corresponding to the additional information, an optimal direction is selected based on the amount of change in the pixel value between the pixels corresponding to each direction among a plurality of preset directions, and the selected optimal direction is used as the filtering direction of the corresponding subblock. And selecting pixels to filter based on the determined filtering direction for each pixel of a boundary of a sub-block, and determining whether to filter the determined filtering direction according to a relationship of pixel values between the selected pixels. The filtering is performed on the selected pixels in the determined filtering direction according to the determined filtering. As described above, the deblocking filter unit 1760 according to the first embodiment may include the deblocking filter unit 190 according to the first embodiment of FIG. 1, that is, the deblocking filtering unit 190 according to the first embodiment of FIG. 2. Compared with the above, only the optimal direction is selected for one pixel corresponding to the additional information transmitted from the encoding apparatus 100 of FIG. 1 without selecting the optimal direction for all pixels of the sub-block boundary, and the other functions are different. All can be configured identically.

The deblocking filter unit 1760 according to the second embodiment of the present invention performs deblocking filtering on the current block, in particular, for each pixel of the boundary of the subblock of the current block, as the additional information of the corresponding subblock. Selecting pixels to filter based on a direction, determining whether to filter the received filtering direction according to a relationship between pixel values between the selected pixels, and selecting the selected pixel of the received filtering direction according to the determined filtering Filter on them. As described above, the deblocking filter unit 1760 according to the second embodiment may include the deblocking filter unit 190 according to the first embodiment of FIG. 1, that is, the deblocking filtering unit 190 according to the first embodiment of FIG. 2. Compared to the above, the process of determining the filtering direction for the corresponding sub-block is omitted and the filtering direction as additional information transmitted from the encoding apparatus 100 of FIG. 1 is used as it is, and all other functions may be configured in the same manner. have.

The deblocking filter unit 1760 according to the third embodiment of the present invention deblocks and filters the current block. In particular, the deblocking filter unit 1760 performs deblocking filtering on the current block. One direction is determined as the filtering direction based on the amount of change in the pixel value between the corresponding pixels, and the second or more selected pixel as the filtering direction is determined as the filtering direction for the second or more pixels on the boundary of the sub block. Alternatively, the filtering direction of the immediately preceding pixel may be determined as the filtering direction, or the filtering direction may be determined according to the filtering direction determination process of the first pixel, and the filtering direction may be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block. Select pixels and compare the determined filtering direction according to the relationship of pixel values between the selected pixels. And determining whether to filter, and performs filtering on the selected pixel of the determined filtering direction, depending on whether the determined filter. As described above, the deblocking filter unit 1740 according to the fourth embodiment includes the deblocking filter unit 190 according to the second embodiment of the encoding apparatus 100 of FIG. 1, that is, the deblocking filter unit 1740 according to the second embodiment of FIG. 3. It may be configured in the same manner as the blocking filtering device 190.

The current block reconstructed by the adder 1750 is combined in a picture unit and output as a reconstructed image, and the current block deblocked and filtered by the deblocking filter 1760 may predict the next picture in the prediction unit 1740. It can be stored as a reference picture used when.

Next, the operation of the apparatus according to the embodiment of the present invention and the deblocking method applicable to the apparatus will be described as described above.

First, the deblocking filtering method of H.264 (“Advanced video coding for generic audiovisual services”, Draft, ITU-T Recommendation H.264, Mar, 2005, p. 182 ~ p. 194) is briefly described.

In the deblocking filtering method of H.264, filtering is performed in a vertical direction as shown in FIG. 18 based on the boundary surface of a discrete cosine transform (DCT) and a quantization unit of a 4x4 block, and then, FIG. 19. Filter in the horizontal direction as shown.

First, before applying the deblocking filter, when there are blocks composed of p0 to p3 and blocks composed of q0 to q3 as shown in FIG. 20, a block boundary strength (BS) value is determined for the boundary of each block, respectively. The strength of the filter boundary is determined based on the block boundary strength value of the filter. The filtering strength is 0 depending on whether the block to which the boundary to be filtered currently belongs uses intra prediction encoding or when distance between the reference blocks for motion compensation is used when inter prediction encoding is used. It is determined by a value between and 4.

If the block boundary strength value is determined, it is determined whether the filtering should be performed by determining whether the pixels of the boundary surface of each block are the boundary surface distorted by the blocking artifact or the edge present in the actual image. In H.264, if it does not satisfy the following [Equation 1], it is determined that the edge exists in the real image, and filtering is not performed. If the condition is satisfied, it is determined that the boundary is distorted by blocking artifacts. Deblocking filter is performed.

[Equation 1]

①

;

;

②

;

;

,

는 양자화 파라미터(Quantization Parameter)에 따라 적응적으로 그 값이 결정되는 값으로써, 현재 경계 면이 블록킹 아티팩트로 발생한 화소 값의 차이인지, 실제 에지에 의한 것인지를 판별할 수 있는 문턱치(Threshold) 값을 의미한다. 양자화 정도가 큰 경우에는 블록킹 아티팩트의 정도가 더욱 크게 발생하므로 문턱치의 값이 크게 설정되는 구조로 되어 있다.here

,

Is a value that is adaptively determined according to a quantization parameter, and is a threshold value for determining whether a current boundary is a difference between pixel values caused by blocking artifacts or an actual edge. it means. If the degree of quantization is large, the degree of blocking artifacts is greater, and thus the threshold value is set large.

When it is determined whether filtering is performed or not, the low band filtering is used according to the block boundary strength value to smooth the pixels on both sides of the boundary surface. The calculation method for filtering can be used, for example, using the method specified in H.264 (“Advanced video coding for generic audiovisual services”, Draft, ITU-T Recommendation H.264, Mar, 2005, p. 182 to p.194). Can be.

As described above, the deblocking filter of H.264 determines the filter strength according to the coding property of the block, and determines whether the edge exists using [Equation 1] on the block boundary surface, while preserving the edge components as much as possible while maintaining the edge components. By adaptively performing deblocking filtering on, we reduce blocking artifacts. However, when the deblocking filtering is performed only in the horizontal and vertical directions as described above, the distortion due to blocking artifacts caused by the discontinuity in the diagonal direction cannot be corrected.

In more detail, when the deblocking filter is performed only in the existing horizontal and vertical directions, p0 to p3 pixels and q0 to q3 when diagonal edge components exist in the block boundary surface as shown in FIG. When [Equation 1] is applied to the boundary surface composed of pixels, the condition is not satisfied and the edge is not determined and the deblocking filtering is not performed. However, if the block boundary strength values of the A blocks and the B blocks are equal to or greater than 1, it can be observed that discontinuities of pixel values occur due to blocking artifacts even though p0 and dq0 have the same value. Therefore, in order to solve the blocking artifacts caused by the discontinuity between the pixels occurring in the diagonal direction, it is necessary to adaptively perform the deblocking filtering in the diagonal direction as well as the vertical and horizontal directions. More specifically, if low-band filtering is performed on the boundary surface in the diagonal direction as shown in (b) of FIG. 21, the pixel values of p0 and q0 are changed and smoothed to solve the distortion caused by the discontinuity of pixel values at the block boundary surface. Could be.

Accordingly, embodiments of the present invention allow the deblocking filter to be performed not only in the vertical and horizontal directions but also in other directions.

In particular, an embodiment of the present invention provides a method for performing deblocking filtering on a pixel-by-pixel basis for various edge directions with a smaller amount of computation since a large amount of computation occurs when performing a deblocking filter having a directionality on a pixel basis.

According to the deblocking filtering method according to the first embodiment of the present invention applicable to the apparatus of FIG. 2, the encoder performs the deblocking filtering by inspecting the edge direction for each pixel, and selects the filtering direction most selected therefrom. In addition, when very little additional information is generated and transmitted to the decoder, the decoder performs directional deblocking filtering using this information. In general, since 4x4 subblocks in an image have generally consistent edge directions, it is reasonable to perform only 1 or 2 directions without performing deblocking filtering on all directions. This method does not change the amount of computation of the encoder, but the advantage of the decoder is that the amount of computation is greatly reduced.

FIG. 22 is a flowchart of a deblocking filtering method according to a first embodiment of the present invention, which is applicable to the apparatus of FIG. 2 and will be described together with the operation of the apparatus.

As shown in FIG. 22, the deblocking filtering method according to the first embodiment of the present invention includes a boundary strength determination step (S2210) for determining an intensity of a boundary between subblocks of a current block; An optimal direction selecting step (S2220) of selecting an optimum direction based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block; A filtering direction determining step (S2230) of determining, as the filtering direction of the corresponding sub block, the most selected direction among the selected plurality of optimal directions for all pixels of the boundary of the sub block; A pixel selection step (S2240) of selecting pixels to be filtered for each pixel at the boundary of the sub block based on the determined filtering direction; A filtering step determining step (S2250) of determining whether to filter in the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering step (S2260) of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering.

The deblocking filtering method according to the first embodiment of the present invention will be described in more detail as follows.

FIG. 4 shows a macroblock consisting of 16x16 pixels, wherein alternating 1, 3, 6, 89, 11, 14, and 16 4x4 blocks of 4x4 subblocks 1-16 are implemented. The deblocking filtering according to the example is applied.

As shown in FIG. 5, the deblocking filter is applied to four boundary surfaces of ①, ②, ③, and ④ of the corresponding subblock to which the deblocking filtering is to be applied. Specific embodiments are as follows. (Steps 1 to 10 are processes performed by the encoder and steps 11 to 13 are processes performed by the decoder.)

Step 1: The boundary strength determiner 210 determines the block boundary strength BS for each boundary of the 4x4 subblock (S2310).

Step 2: If BS is 0, no deblocking filtering is performed, and if BS is 1 to 4, the conditional expression is examined to determine whether deblocking filtering should be performed.

Step 3: The filtering direction determiner 230 may select the pixel boundaries 1 to 16 of the 4x4 subblock boundary surfaces ①, ②, ③, and ④ as shown in FIG. 6 from among nine preset directions as shown in FIG. 7. The most suitable deblocking filtering direction is determined (S2220).

Step 4: The filtering direction determiner 230 obtains a change amount of pixel values between pixels corresponding to nine preset directions on the boundary surface of pixels 1 to 16 in order to obtain the most suitable direction, and among the nine directions The direction in which the change amount is the smallest is the edge direction, and the optimal direction is selected as the best direction for deblocking filtering (S2220).

는 화소 경계 면의 화소(pixel)를 기준으로 각 방향에 따라 선택된 방향 세트에서의 화소들간의 화소값의 변화량을 나타낸다.

,

은 가중치(weight)를 나타내는 상수로써, 블록 간의 경계 근처에 있는 화소 간의 변화량이 경계에서 멀리 떨어져 있는 화소 간의 변화량보다 전체 변화량에 더 큰 비중을 갖도록

,

을 만족하는 값으로 설정된다(S2220).Step 5: The filtering direction determiner 230 calculates a change amount between pixels in each of nine directions at the pixel boundary, as follows.

Denotes an amount of change in pixel values between pixels in a direction set selected according to each direction on the basis of a pixel on a pixel boundary surface.

,

Is a constant representing the weight, so that the amount of change between pixels near the boundary between blocks has a greater proportion to the total amount of change than the amount of change between pixels far away from the boundary.

,

It is set to a value satisfying (S2220).

Step 6: In the filtering direction determiner 230, an equation for obtaining variation amounts for each of nine directions is specifically as follows (S2220).

The change amount with respect to the 90 degree direction shown in FIG. 8 is calculated | required as following formula (2).

[Equation 2]

The change amount with respect to the 78.75 degree direction shown in FIG. 9 is calculated | required as following [Equation 3].

[Equation 3]

The change amount with respect to the 67.5 degree direction shown in FIG. 10 is calculated | required as following [Equation 4].

[Equation 4]

The change amount with respect to the 56.25 degree direction shown in FIG. 11 is calculated | required as following [Equation 5].

[Equation 5]

The change amount with respect to the 45 degree direction shown in FIG. 12 is calculated | required as following [Equation 6].

[Equation 6]

The change amount with respect to the direction of 101.25 degree shown in FIG. 13 is calculated | required as following [Equation 7].

[Equation 7]

The change amount with respect to the 112.5 degree direction shown in FIG. 14 is calculated | required as following [Equation 8].

[Equation 8]

The change amount with respect to the 123.75 degree direction shown in FIG. 15 is calculated | required as following [Equation 9].

[Equation 9]

The change amount with respect to the 135 degree direction shown in FIG. 16 is calculated | required as following [Equation 10].

[Equation 10]

Step 7: Next, the filtering direction determiner 230 determines an optimal direction, which is most selected among the boundaries between pixels 1 to 16 of FIG. 6, as the deblocking filtering direction of the corresponding subblock (S2230).

Step 8: Next, the filtering unit 250 determines whether deblocking filtering is performed in the filtering direction determined above with respect to each boundary between 1 to 16 pixels of FIG. 6. In this case, a conditional expression similar to that of [Equation 1] is used. In the present embodiment, since nine preset filtering directions are used in total, it is determined whether filtering is performed using a separate conditional expression for each of the nine directions, and when the conditional expression is satisfied, Deblocking filtering is performed in the filtering direction. If the conditional expression is not satisfied, the conditional expression is applied to the vertical direction of the boundary surface, and if it is satisfied, deblocking filtering is performed in the vertical direction of the boundary surface (S2240 to S2260).

,

는 상수로 문턱치 값을 조절하여, 각각의 방향에서 디블록킹 필터링의 수행 정도를 결정하는 것이 가능하다. Step 9: The pixels required for the calculation in order to apply the conditional expression in each of the nine directions are as shown in Figs. 8 to 16 (S2240). The conditional expression for each direction is specifically as follows. At this time,

,

It is possible to determine the degree of performance of deblocking filtering in each direction by adjusting the threshold value with a constant.

The conditional expression for the 90 degree direction of FIG. 8 is as follows [Condition 1].

[Condition 1]

;

;

;

;

The conditional expression for the 78.75 degree direction of FIG. 9 is as follows [conditional expression 2].

[Condition Formula 2]

;

;

;

;

The conditional formula for the direction of 67.5 degrees in FIG. 10 is as follows [Condition 3].

[Condition 3]

;

;

;

;

The conditional expression for the 56.25 degree direction of FIG. 11 is as follows.

[Condition 4]

;

;

;

;

The conditional expression for the 45 degree direction of FIG. 12 is as follows [conditional expression 5].

[Condition 5]

;

;

;

;

The conditional expression for the direction of 101.25 degree of FIG. 13 is as follows [Condition 6].

[Condition 6]

;

;

;

;

The conditional expression for the direction of 112.5 degrees in FIG. 14 is as follows [Condition 7].

[Condition 7]

;

;

;

;

The conditional expression for the direction of 123.75 degree of FIG. 15 is as follows.

[Condition 8]

;

;

;

;

The conditional expression for the 135 degree direction of FIG. 16 is as follows [conditional expression 9].

[Condition 9]

;

;

;

;

Step 10: On the other hand, one of the pixels performed in the deblocking filtering direction (deblocking filtering direction determined in step 7) of the corresponding sub-block rather than the vertical direction is selected, and one-bit additional information is allocated to the pixel. Additional information is sent from the encoder to the decoder. As another example, the deblocking filtering direction (deblocking filtering direction determined in step 7) of the corresponding subblock may be recorded and stored as additional information, and the additional information may be transmitted from the encoder to the decoder.

Step 11: The decoder selects the pixel at the position designated by the encoder by using the additional information received from the encoder. The same method as used in the encoder is used to find a direction to perform the deblocking filtering of the pixel. This direction becomes the deblocking filtering direction of the corresponding 4x4 subblock (set to the same filtering direction found in step 7).

Step 12: When the filtering direction of the corresponding subblock is determined, the decoder performs deblocking filtering by performing the same process as steps 8 to 9, similarly to the encoder.

Step 13: In the case of applying the above method, the decoder determines a direction for one pixel and determines that direction is the main direction of the current sub-block, without having to determine a direction in which to perform deblocking filtering on each pixel, unlike an encoder. Deblocking filtering reduces the amount of computation. However, some additional information must be transmitted from the encoder to the decoder.

FIG. 23 is a flowchart of a deblocking filtering method according to a second embodiment of the present invention. Since it is applicable to the apparatus of FIG. 3, it will be described together with the operation of the apparatus.

As shown in FIG. 23, the deblocking filtering method according to the second embodiment of the present invention includes a boundary strength determination step (S2310) for determining an intensity of a boundary between subblocks of a current block; A first filtering direction determining step of determining one direction as a filtering direction with respect to the first pixel of the boundary of the sub-block, based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions ( S2320); For the second or more pixels on the boundary of the sub-block, the direction most selected so far for the current sub-block is determined as the filtering direction, or the filtering direction of the immediately preceding pixel is determined as the optimal direction, or the filtering direction of the first pixel is determined. Determining a filtering direction according to the determining process (S2330); A pixel selecting step (S2340) of selecting pixels to be filtered for each pixel of the boundary of the sub block based on the determined filtering direction; A filtering step determining step (S2350) of determining whether to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And a filtering step (S2360) of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering.

The deblocking filtering method according to the second embodiment of the present invention will be described in more detail as follows.

The deblocking filtering method according to the second embodiment of the present invention is a method of performing deblocking filtering on a pixel-by-pixel basis while reducing the amount of computation compared to the conventional method without the need for transmitting additional information from the encoder to the decoder.

FIG. 4 shows a macroblock consisting of 16x16 pixels, wherein alternating 1, 3, 6, 89, 11, 14, and 16th 4x4 subblocks of 4x4 subblocks 1 to 16 are shown in FIG. The deblocking filtering according to the embodiment is applied. As shown in FIG. 5, the deblocking filter is applied to four boundary surfaces of ①, ②, ③, and ④ of the corresponding subblock to which the deblocking filtering is to be applied. Specific embodiments are as follows.

Step 1: The boundary strength determiner 310 determines the block boundary strength BS for each boundary of the 4x4 block (S2310).

Step 2: If BS is 0, no deblocking filtering is performed, and if BS is 1 to 4, the conditional expression is examined to determine whether deblocking filtering should be performed.

Step 3: The filtering direction determiner 330 examines a direction in which the deblocking filtering is to be applied from among nine preset directions for the pixel boundary of the ① boundary surface of the corresponding subblock of FIG. 6. The filtering direction determiner 330 obtains a change amount of pixel values between pixels for nine preset directions as in the deblocking filtering method according to the first embodiment of FIG. Is determined as the first filtering direction. The amount of change in each of the nine directions is obtained by applying [Equation 2] to [Equation 10] (S2320). The filtering unit 350 performs deblocking filtering in the determined first filtering direction when the conditional expression corresponding to the determined first filtering direction is satisfied (see conditional expressions 1 to 9 in step 9 of the first embodiment of FIG. 22). Otherwise, deblocking filtering is not performed (S2340-S2360).

Step 4: Next, the filtering direction determiner 330 determines, as the filtering direction, the first filtering direction used at the pixel boundary 1 at the pixel boundary 2 of the ① boundary surface of FIG. 6, and the filtering unit 350 determines the determined filtering direction. It is determined whether deblocking filtering should be performed in the filtering direction (S2330-S2350). At this time, the filtering unit 350 performs deblocking filtering if the conditional expression is satisfied using the corresponding conditional expressions of the conditional expressions 1 to 9 of step 9 of the first embodiment of FIG. The direction determiner 330 again checks the directions of the nine directions as in the first pixel boundary to determine the direction with the least amount of change as the filtering direction (S2330), and then deblocking filtering through the filter 350. It performs (S2340-S2360).

Step 5: Next, the filtering direction determiner 330 determines the filtering direction most selected so far on the pixel boundary surface of interface ① in FIG. 6 as the filtering direction, and the filter 350 determines the determined filtering direction. It is determined whether deblocking filtering should be performed by applying the corresponding conditional expressions of Conditional Expressions 1 to 9 with respect to the direction (S2330-S2350), and if the conditional expression is satisfied, deblocking filtering is performed (S2360). If it is not satisfied, the filtering direction determiner 330 determines the filtering direction (the filtering direction of the pixel boundary of pixel 2) that was used immediately before as the filtering direction, and the filtering unit 350 applies the conditional expression to the determined filtering direction. Deblocking filtering is performed (S2330-S2350). If the conditional expression is satisfied, deblocking filtering is performed (S2360). If it is not satisfied again, the filtering direction determiner 330 again checks the new directions for the nine directions as in the above process for the pixel boundary 1 and determines the direction with the smallest change as the filtering direction (S2330). Deblocking filtering is performed through the filtering unit 350 in operation S2340-S2360.

Step 6: As described above, for more than two pixel boundaries at the boundary of each pixel of the boundary of the 4x4 subblock, it is first checked whether deblocking filtering is applied using the direction most selected in the subblock so far. Filtering is performed in that direction, and if it is not satisfied, it checks whether the deblocking filtering is applied to the direction used for the boundary between pixels immediately before, and performs filtering in that direction. Filtering may be performed by determining a direction of deblocking filtering by obtaining a change amount between boundaries for each direction.

According to the method of the second embodiment described above, the amount of computation is reduced because it is not necessary to check the other direction first if the condition is satisfied first after checking the direction most likely for each pixel boundary in the 4x4 subblock.

The deblocking filtering method according to an embodiment of the present invention may be implemented as computer readable codes on a computer readable recording medium. Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which are also implemented in the form of carrier waves (for example, transmission over the Internet). Include. 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.

In the above description, all elements constituting the embodiments of the present invention are described as being operated in combination, and the present invention is necessarily limited to these embodiments. In other words, within the scope of the present invention, all of the components may be selectively operated in combination with one or more. In addition, although all of the components may be implemented in one independent hardware, each or all of the components may be selectively combined to perform some or all functions combined in one or a plurality of hardware. It may be implemented as a computer program having a. Codes and code segments constituting the computer program may be easily inferred by those skilled in the art. Such a computer program may be stored in a computer readable storage medium and read and executed by a computer, thereby implementing embodiments of the present invention. The storage medium of the computer program may include a magnetic recording medium, an optical recording medium, a carrier wave medium, and the like.

In addition, the terms "comprise", "comprise" or "having" described above mean that the corresponding component may be included, unless otherwise stated, and thus excludes other components. It should be construed that it may further include other components instead. All terms, including technical and scientific terms, have the same meanings as commonly understood by one of ordinary skill in the art unless otherwise defined. Terms commonly used, such as terms defined in the dictionary, should be interpreted to coincide with the contextual meaning of the related art, and shall not be construed in an ideal or overly formal sense unless explicitly defined in the present invention.

The above description is merely illustrative of the technical idea of the present invention, and those skilled in the art to which the present invention pertains may make various modifications and changes without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention but to describe the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

As described above, the present invention is applied to a video encoding and decoding technology field, and proposes a method for applying a deblocking filter on a pixel-by-pixel basis for not only a vertical horizontal direction but also a plurality of other directions, thereby providing a plurality of diagonal directions. It is a very useful invention that improves the image quality by effectively eliminating the block distortion caused by the discontinuity in the diagonal direction caused by the edge to, and effectively reduces the amount of computation generated when the block distortion occurs.

1 is a block diagram of a video encoding apparatus according to an embodiment of the present invention;

2 is a block diagram of a deblocking filtering device according to a first embodiment of the present invention;

3 is a block diagram of a deblocking filtering device according to a second embodiment of the present invention;

4 is a diagram illustrating a sub block in a macro block applied to an embodiment of the present invention;

5 is a diagram illustrating a boundary of a sub block applied to an embodiment of the present invention;

6 is a diagram illustrating pixel boundaries of sub-block boundaries applied to an embodiment of the present invention;

7 illustrates an example of a filtering direction according to an embodiment of the present invention;

8 to 16 illustrate examples of pixels selected according to the filtering direction of FIG. 7;

17 is a block diagram of a video decoding apparatus according to an embodiment of the present invention;

18 and 19 are views showing the boundaries of the macro block in the vertical and horizontal directions,

20 illustrates an example of pixels to be horizontally and vertically filtered in the case of vertical and horizontal block boundaries;

21 is a diagram illustrating a case of filtering in a vertical direction when an edge in a diagonal direction exists, (b) is a diagram illustrating a filter in a diagonal direction;

22 is a flowchart of a deblocking filtering method according to a first embodiment of the present invention;

23 is a flowchart of a deblocking filtering method according to a second embodiment of the present invention.

<Description of Symbols for Main Parts of Drawings>

100: encoding device (or coder)

110: prediction unit 120: subtraction unit

130: transform unit 140: quantization unit

150: encoding unit 160: inverse quantization unit

170: inverse transform unit 180: adder

190: deblocking filter unit, adaptive deblocking filtering device

210,310: boundary strength determiner 230,330: filtering direction determiner

250,350: filtering unit

1700: decoding device (or decoder)

1710: decoding unit 1720: inverse quantization unit

1730: inverse transform unit 1740: predictor

1750: adder 1760: deblocking filter part

Claims (27)

A boundary strength determiner for determining the strength of the boundary between sub-blocks of the current block; An optimal direction may be selected based on the amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block, and the predetermined direction may be selected for all pixels of the boundary of the sub-block. A filtering direction determiner configured to determine, as a filtering direction of the corresponding sub-block, the most selected direction among a plurality of optimal directions; And Selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and determining whether to filter the determined filtering direction according to a relationship of pixel values between the selected pixels, and determining the filtering A filtering unit which performs filtering on the selected pixels in the determined filtering direction according to whether or not; Deblocking filtering device comprising a. The method of claim 1, The filtering direction determiner, Absolute value of the pixel value difference between the first pixels facing the corresponding boundary, absolute value of the pixel value difference between the first pixel and the second pixel on the basis of the corresponding boundary, and based on the corresponding boundary Absolute value of the pixel value difference between the second pixel and the first pixel on the side, absolute value of the pixel value difference between the third pixel and the second pixel on the basis of the corresponding boundary, and other side based on the corresponding boundary And calculating the amount of change based on the absolute value of the difference in pixel values between the third pixel and the second pixel, and selecting a direction in which the amount of change is the smallest as the optimum direction. The method of claim 1, The filtering unit, The absolute value of the pixel value difference between the first pixels facing the boundary, and the relationship of the pixel value between the selected pixels, the pixel value difference between the second pixel and the first pixel of one side based on the boundary And filtering in the determined filtering direction when the absolute value and the absolute value of the pixel value difference between the second pixel and the first pixel on the other side are smaller than predetermined thresholds, respectively, based on the boundary. Deblocking filtering device. The method of claim 3, wherein The filtering unit, And if the determined filtering direction does not satisfy the condition, check whether the condition is satisfied with respect to the vertical direction of the corresponding boundary, and perform filtering according to the checking result. The method of claim 1, The predetermined plurality of directions include 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees and 135 degrees. The method of claim 1, The filtering direction determiner, And recording the determined filtering direction as additional information for the corresponding subblock or by allocating additional information for pixels filtered in the determined filtering direction among pixels on the boundary of the corresponding subblock. A boundary strength determiner for determining the strength of the boundary between sub-blocks of the current block; For the first pixel of the boundary of the sub-block, one direction is determined as the filtering direction based on the amount of change in the pixel value between the pixels corresponding to each direction among a plurality of preset directions, and two of the boundaries of the sub-block are determined. For the second or more pixels, the filtering direction is determined as the filtering direction for the current subblock, the filtering direction of the immediately preceding pixel is determined as the filtering direction, or the filtering direction is determined according to the filtering direction determination process of the first pixel. A filtering direction determiner; And Selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and determining whether to filter the determined filtering direction according to a relationship of pixel values between the selected pixels, A filtering unit configured to perform filtering on the selected pixels in the determined filtering direction according to filtering; Deblocking filtering device comprising a. The method of claim 7, wherein The filtering direction determiner, Absolute value of the pixel value difference between the first pixels facing the corresponding boundary, absolute value of the pixel value difference between the first pixel and the second pixel on the basis of the corresponding boundary, and based on the corresponding boundary Absolute value of the pixel value difference between the second pixel and the first pixel on the side, absolute value of the pixel value difference between the third pixel and the second pixel on the basis of the corresponding boundary, and other side based on the corresponding boundary And calculating the change amount based on an absolute value of a difference between pixel values of a third pixel and a second pixel of the second pixel, and determining a direction in which the change amount is the smallest as the filtering direction. The method of claim 7, wherein The filtering unit, The absolute value of the pixel value difference between the first pixels facing the boundary, and the relationship of the pixel value between the selected pixels, the pixel value difference between the second pixel and the first pixel of one side based on the boundary And filtering in the determined filtering direction when the absolute value and the absolute value of the pixel value difference between the second pixel and the first pixel on the other side are smaller than predetermined thresholds, respectively, based on the boundary. Deblocking filtering device. The method of claim 7, wherein The predetermined plurality of directions include 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees and 135 degrees. A boundary strength determining step of determining the strength of a boundary between sub-blocks of the current block; An optimal direction selecting step of selecting an optimum direction based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block; A filtering direction determining step of determining, as the filtering direction of the corresponding sub block, the most selected direction among the selected plurality of optimal directions for all pixels on the boundary of the sub block; A pixel selecting step of selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering step determining step of determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And A filtering step of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering; Deblocking filtering method comprising a. The method of claim 11, The amount of change in the optimum direction selection step, Absolute value of the pixel value difference between the first pixels facing the corresponding boundary, absolute value of the pixel value difference between the first pixel and the second pixel on the basis of the corresponding boundary, and based on the corresponding boundary Absolute value of the pixel value difference between the second pixel and the first pixel on the side, absolute value of the pixel value difference between the third pixel and the second pixel on the basis of the corresponding boundary, and other side based on the corresponding boundary The deblocking filtering method is calculated based on the absolute value of the difference in pixel values between the third pixel and the second pixel, and the direction in which the amount of change is the smallest is selected as the optimal direction. The method of claim 11, The determining whether to filter, The absolute value of the pixel value difference between the first pixels facing the boundary based on the relationship between the pixel values between the selected pixels, and the pixel value difference between the second pixel and the first pixel on one side based on the corresponding boundary. And whether or not the filtering is performed according to whether or not the absolute value of and the absolute value of the pixel value difference between the second pixel and the first pixel on the other side based on the corresponding boundary satisfy a condition smaller than the preset thresholds, respectively. Deblocking filtering method characterized in that. The method of claim 13, The determining whether to filter, If the determined filtering direction does not satisfy the condition, it is determined whether the condition is satisfied with respect to the vertical direction of the corresponding boundary, and according to the result of the checking, it is determined whether to filter the vertical direction. Deblocking filtering method. The method of claim 11, The predetermined plurality of directions include 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees, and 135 degrees. The method of claim 11, In the determining of the filtering direction, recording the determined filtering direction as additional information for the corresponding subblock, or allocating and recording additional information for pixels filtered in the determined filtering direction among pixels on the boundary of the corresponding subblock. Deblocking filtering method comprising a. A boundary strength determining step of determining the strength of a boundary between sub-blocks of the current block; A first filtering direction determining step of determining, with respect to the first pixel of the boundary of the sub block, one direction as a filtering direction based on an amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions; For the second or more pixels on the boundary of the sub-block, the direction most selected so far for the current sub-block is determined as the filtering direction, or the filtering direction of the immediately preceding pixel is determined as the optimal direction, or the filtering direction of the first pixel is determined. Determining a filtering direction according to the determining process; A pixel selecting step of selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering step determining step of determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And A filtering step of performing filtering on the selected pixels in the determined filtering direction according to the determined filtering; Deblocking filtering method comprising a. The method of claim 17, The amount of change in the first filtering direction determination step, Absolute value of the pixel value difference between the first pixels facing the corresponding boundary, absolute value of the pixel value difference between the first pixel and the second pixel on the basis of the corresponding boundary, and based on the corresponding boundary Absolute value of the pixel value difference between the second pixel and the first pixel on the side, absolute value of the pixel value difference between the third pixel and the second pixel on the basis of the corresponding boundary, and other side based on the corresponding boundary The deblocking filtering method is calculated based on the absolute value of the difference in pixel values between the third pixel and the second pixel, and a direction in which the calculated change amount is the smallest is selected as the filtering direction. The method of claim 17, The determining whether to filter, The absolute value of the pixel value difference between the first pixels facing the boundary based on the relationship between the pixel values between the selected pixels, and the pixel value difference between the second pixel and the first pixel on one side based on the corresponding boundary. And whether or not the filtering is performed according to whether or not the absolute value of and the absolute value of the pixel value difference between the second pixel and the first pixel on the other side with respect to the corresponding boundary satisfy a condition smaller than the preset thresholds, respectively. Deblocking filtering method characterized in that. The method of claim 17, The set plurality of directions include 45 degrees, 56.25 degrees, 67.5 degrees, 78.75 degrees, 90 degrees, 101.25 degrees, 112.5 degrees, 123.75 degrees, and 135 degrees. A prediction unit for predicting a current block of an image and generating a prediction block; A subtraction unit for generating a residual block by subtracting the prediction block from the current block; A transformer for converting the residual block into a frequency domain; A quantizer for quantizing the transformed residual block; An encoder which encodes the quantized residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block into a residual block having a pixel signal on a time axis; An adder configured to add the prediction block to the inverse transformed residual block to restore a current block; And The deblocking filtering of the restored current block is performed, and an optimal direction is determined based on the amount of change of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the subblock of the restored current block. Select a direction among the selected plurality of optimal directions for all the pixels on the boundary of the sub block as the filtering direction of the corresponding sub block, and determine the determined filtering direction for each pixel on the boundary of the sub block. Select the pixels to be filtered based on the control, determine whether to filter in the determined filtering direction according to the relationship between the pixel values between the selected pixels, and determine the selected pixels in the determined filtering direction according to the determined filtering. A deblocking filter unit performing filtering; Video encoding apparatus comprising a. A prediction unit for predicting a current block of an image and generating a prediction block; A subtraction unit for generating a residual block by subtracting the prediction block from the current block; A transformer for converting the residual block into a frequency domain; A quantizer for quantizing the transformed residual block; An encoder which encodes the quantized residual block; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block into a residual block having a pixel signal on a time axis; An adder configured to add the prediction block to the inverse transformed residual block to restore a current block; And The deblocking filtering of the restored current block is performed, and the first pixel of the boundary of the subblock of the restored current block is one based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions. Determine the direction of as the filtering direction, and for the second or more pixels on the boundary of the sub block, determine the direction most selected so far for the current sub block as the filtering direction, or determine the filtering direction of the immediately preceding pixel as the filtering direction, or And determining a filtering direction according to the filtering direction determining process of the first pixel, selecting pixels to be filtered based on the determined filtering direction for each pixel of the boundary of the sub-block, and selecting a pixel value between the selected pixels. Determine whether to filter the determined filtering direction according to the relationship; The deblocking filter unit according to whether the specified filter that performs filtering on the selected pixel of the determined filtering direction; Video encoding apparatus comprising a. A decoder which extracts a residual block by decoding the bitstream; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And Deblocking and filtering the reconstructed current block, and among the pixels corresponding to each of a plurality of preset directions for a specific pixel corresponding to the received additional information among all the pixels on the boundary of the sub-block of the reconstructed current block. Selecting an optimal direction based on the amount of change in the pixel value, determining the selected optimal direction as the filtering direction of the corresponding subblock, and filtering the pixel based on the determined filtering direction for each pixel at the boundary of the subblock. Deblocking for filtering the selected filtering direction according to the relationship between the pixel values of the selected pixels, and performing filtering on the selected pixels in the determined filtering direction according to the determined filtering. Filter unit; Video decoding apparatus comprising a. A decoder which extracts a residual block by decoding the bitstream; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And Deblocking and filtering the current block reconstructed by the adder, and selecting pixels to be filtered based on the filtering direction received as additional information of the corresponding subblock, for each pixel of the boundary of the subblock of the reconstructed current block. And determining whether to filter the received filtering direction according to the relationship between the pixel values between the selected pixels, and performing deblocking on the selected pixels of the received filtering direction according to the determined filtering. Filter unit; Video encoding apparatus comprising a. A decoder which extracts a residual block by decoding the bitstream; An inverse quantizer for inversely quantizing the residual block; An inverse transform unit for inversely transforming the inverse quantized residual block; A prediction unit for predicting a current block to generate a prediction block; An adder configured to add the inverse transformed residual block and the prediction block to restore the current block; And Deblocking filtering the current block reconstructed by the adder, and for the first pixel of the boundary of the subblock of the reconstructed current block, an amount of change in pixel values between pixels corresponding to each direction among a plurality of preset directions. Determine a direction based on the filtering direction, and for the second or more pixels on the boundary of the sub block, determine the direction most selected so far as the filtering direction for the current sub block, or filter the filtering direction of the immediately preceding pixel. Or the filtering direction according to the filtering direction determination process of the first pixel, and selecting pixels to be filtered on the basis of the determined filtering direction for each pixel of the boundary of the sub-block, and between the selected pixels. Whether to filter for the determined filtering direction according to the relationship of pixel values of A deblocking filter unit configured to determine and to filter the selected pixels in the determined filtering direction according to the determined filtering; Video decoding apparatus comprising a. A boundary strength determination function for determining the strength of a boundary between sub-blocks of the current block; An optimum direction selection function for selecting an optimum direction based on a change amount of pixel values between pixels corresponding to each direction among a plurality of preset directions for each pixel of the boundary of the sub-block; A filtering direction determining function for determining, as the filtering direction of the corresponding sub block, the most selected direction among the plurality of selected optimal directions for all pixels on the boundary of the sub block; A pixel selection function for selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering or not determining function for determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And A filtering function for performing filtering on the selected pixels in the determined filtering direction according to the determined filtering; And a computer readable recording medium having recorded thereon a program. A boundary strength determination function for determining the strength of a boundary between sub-blocks of the current block; A first filtering direction determining function for determining the first direction as the filtering direction for the first pixel of the boundary of the sub-block, based on the amount of change in the pixel value between the pixels corresponding to each direction among a plurality of preset directions. ; For the second or more pixels on the boundary of the sub-block, the direction most selected so far for the current sub-block is determined as the filtering direction, or the filtering direction of the immediately preceding pixel is determined as the optimal direction, or the filtering direction of the first pixel is determined. A second filtering direction determining function for determining a filtering direction according to the determining process; A pixel selection function for selecting pixels to be filtered for each pixel at a boundary of the sub block based on the determined filtering direction; A filtering or not determining function for determining whether or not to filter the determined filtering direction according to a relationship between pixel values between the selected pixels; And A filtering function for performing filtering on the selected pixels in the determined filtering direction according to the determined filtering; And a computer readable recording medium having recorded thereon a program.

Images