KR100982514B1 - Method and apparatus for predictive intra coding for image data - Google Patents

Abstract

Disclosed are a method and an apparatus for intra prediction encoding of an image. In the intra prediction encoding method of an image according to the present invention, intra prediction encoding is performed on an input block in units of sub blocks having a predetermined size, and among the plurality of sub block prediction modes, for each sub block included in the input block. Determining an intra prediction mode; And when the frequency of the prediction mode having the highest frequency of occurrence among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is used for intra prediction encoding of the input block. Determining the final prediction mode. According to the present invention, the intra prediction mode of an image can be determined at a higher speed by reducing the amount of computation required to determine the intra prediction mode without degrading the intra prediction encoding performance.

Description

Intra predictive coding method for video and its apparatus {Method and apparatus for predictive intra coding for image data}

1 (a) to (i) are diagrams illustrating an intra 4x4 mode for intra prediction encoding of H.264,

2A to 2D are diagrams illustrating an intra 16x16 mode for intra prediction encoding of H.264,

3 is a diagram illustrating an example of a block diagram of a video encoding apparatus according to the present invention;

4 is a diagram illustrating an example of a block diagram of an intra prediction encoding apparatus according to the present invention;

5 is a diagram illustrating an example of a flowchart of an intra prediction encoding method according to the present invention.

The present invention relates to intra prediction encoding of an image, and more particularly, to an intra prediction encoding method and an apparatus of an image capable of determining an intra prediction mode for intra prediction encoding of an image at a higher speed.

In H.264 / MPEG-4 Advanced Video Coding (AVC), one of the compression standards for video signals, when intra coding a current macro block, encoding is performed through spatial prediction using neighboring already encoded pixel values. Perform In this case, the intra mode is largely divided into an intra 4x4 mode and an intra 16x16 mode. The intra 4x4 mode is further divided into nine modes as shown in FIGS. 1A to 1I according to the prediction direction, and the intra 16x16 mode is shown in FIGS. 2A to 2D as shown in FIGS. It is divided into four modes.

For example, according to the vertical 4x4 mode illustrated in FIG. 1A, when the prediction encoding operation of the current block of 4x4 is described, the pixel values of the pixels A to D adjacent to the upper side of the 4x4 current block are 4x4. Predict the pixel value of the current block. That is, the value of pixel A is predicted by four pixel values included in the first column of the 4x4 current block, the value of pixel B is predicted by four pixel values included in the second column of the 4x4 current block, and the value of pixel C is 4x4. The pixel value is predicted by four pixel values included in the third column of the current block, and the pixel D value is predicted by four pixel values included in the fourth column of the 4x4 current block. After that, the difference value is obtained by subtracting the actual value and the predicted value of the pixel included in the 4x4 current block, and then encoding the difference value.

The H.264 encoder encodes a current macroblock in a total of 13 modes of the above-described intra 4x4 mode and intra 16x16 mode, and then performs intra coding in a mode having the least cost among them. In more detail, first, four intra 16x16 prediction modes are performed on the current macroblock to select a mode having the lowest cost. Thereafter, nine intra 4x4 prediction modes are sequentially performed on 16 subblocks in units of 4x4 subblocks to select a mode having the lowest cost for each subblock. Finally, the cost of the intra 16x16 prediction mode is compared with the cost of the intra 4x4 prediction mode, which is the sum of the costs of the 16 subblocks, and finally the mode having the lowest cost is selected.

The cost is a function value that represents the magnitude of the accuracy of prediction encoding and the amount of bits generated. Functions for measuring costs include sum of absolute difference (SAD), sum of absolute transformed difference (SATD), sum of squared difference (SSD), mean of absolute difference (MAD), or Lagrange function (LAD). have.

When the actual video is encoded in H.264, the selection ratio of the intra 16x16 prediction mode is significantly smaller than that of the intra 4x4 prediction mode. 70-80% of the macroblocks in the I picture that intra-encode all macroblocks included in the picture are selected to intra 4x4 mode. The reason for introducing the intra 16x16 prediction mode is to effectively encode a very flat (smooth) region in the image, and a very flat (smooth) region in the general image is a part of the whole image.

However, according to the related art, encoding is always performed for all macro blocks to be intra-coded in all 13 modes of 9 4x4 modes and 4 intra 16x16 modes, and the mode having the least cost among them is selected. Decisions are not efficient in terms of computation.

Accordingly, an aspect of the present invention is to provide an intra prediction encoding method and apparatus for an image capable of determining an intra prediction mode for intra prediction encoding of an image more simply and at higher speed.

In order to achieve the above object, the intra prediction encoding method of an image according to an aspect of the present invention,

Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And when the frequency of the prediction mode having the highest frequency of occurrence among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is used for intra prediction encoding of the input block. Determining the final prediction mode.

In order to achieve the above object, the intra prediction encoding method of an image according to another aspect of the present invention,

Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And if the number of different prediction modes among the plurality of intra prediction modes determined for each sub block is greater than a predetermined threshold Th _b , the intra prediction mode for each sub block may be used as a final result for intra prediction encoding of the input block. Determining the prediction mode.

In order to achieve the above object, the intra prediction encoding apparatus for an image according to an aspect of the present invention,

An intra prediction encoder configured to perform intra prediction encoding on an input block in units of a predetermined size, and to determine an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And when the frequency of the prediction mode having the highest frequency of occurrence among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is used for intra prediction encoding of the input block. It includes a prediction mode determiner for determining the final prediction mode.

In order to achieve the above object, the intra prediction encoding apparatus for an image according to another aspect of the present invention,

An intra prediction encoder configured to perform intra prediction encoding on an input block in units of a predetermined size, and to determine an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And if the number of different prediction modes among the plurality of intra prediction modes determined for each sub block is greater than a predetermined threshold Th _b , the final prediction for intra prediction encoding of the input block is performed. It includes a prediction mode determiner that determines the mode.

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

In the present invention, the intra 16x16 prediction process is skipped using the intra 4x4 prediction result without reducing the encoding performance, thereby reducing the amount of computation required for intra coding.

3 is a diagram illustrating an example of a block diagram of a video encoding apparatus according to the present invention. Referring to FIG. 3, the video encoding apparatus includes a transform / quantizer 110, a dequantizer / inverse transform 131, a deblocking filter 133, a picture reproducing unit 135, a motion compensated predictor 137, and an intra prediction unit 139. , A motion estimator 150, a subtractor 170, and an entropy coding unit 190.

Image data is input to a video encoding apparatus in units of macroblocks of 16x16 pixels. The Transform / Quantizer 110 transforms the input macroblock according to a predetermined method and then quantizes it. A representative image transformation technique is DCT (Discrete Cosing Transform).

The dequantizer / Inverse transform 131 receives inverse quantized image data after transformation from the Transform / Quantizer 110 and inverse quantizes and inverse transforms. The deblocking filter 133 receives inverse quantized and inverse transformed image data from the dequantizer / inverse transform 131 and performs filtering to remove a blocking effect.                     

The picture reproducing unit 135 receives the filtered image data from the Deblocking filter 133 and reproduces and stores the image in picture units. The picture is an image in a frame unit or an image in a field unit. The picture reproducing unit 135 includes a buffer (not shown) capable of storing a plurality of pictures. The plurality of pictures stored in the buffer are pictures provided for motion estimation and referred to as reference pictures.

The motion estimator 150 receives at least one reference picture stored in the picture reproducing unit 135 and performs motion estimation of an input macro block to output motion data including a motion vector, an index indicating a reference picture, and a block mode.

The motion compensated predictor 137 is a macroblock corresponding to an input macroblock from a reference picture used for motion estimation among a plurality of reference pictures stored in the picture reproducing unit 135 according to the motion data input from the motion estimator 150. Extract and print

The subtractor 170 receives a macroblock in a reference picture corresponding to the input macroblock from the motion compensated predictor 137 and performs a difference operation with the input macroblock when performing prediction encoding on the input macroblock between pictures. Output the (residue signal).

The residual signal output from the subtractor 170 is transformed and quantized by the Transform / Quantizer 110 and entropy coded by the Entropy coding unit 190 to generate an output bit stream.

The intra prediction unit 139 performs intra prediction encoding according to the present invention.                     

Meanwhile, the video decoder 130 that decodes the bit stream generated by the video encoding apparatus described above may include a dequantizer / inverse transform 131, a deblocking filter 133, a picture reproducing unit 135, and a motion compensated predictor ( 137 and an intra prediction unit 139.

4 is a diagram illustrating an example of a block diagram of an intra prediction encoding apparatus according to the present invention. The intra prediction encoding apparatus shown in FIG. 4 corresponds to the intra prediction unit 139 shown in FIG. 3. Referring back to FIG. 4, the intra prediction encoding apparatus includes an intra prediction encoding unit 139a and a prediction mode determination unit 139b.

The intra prediction encoder 139a receives the current macroblock to be predictively encoded and performs intra prediction encoding in an intra 4x4 mode or an intra 16x16 mode as shown in FIGS. 1A to 1I.

The prediction mode determiner 139b receives the prediction mode or the cost of each subblock or the current macroblock from the intra prediction encoder 139a and determines the final prediction mode for the prediction encoding of the current macroblock.

The intra prediction encoding method according to the present invention will be described based on the structure of the intra prediction encoding apparatus according to the present invention described above.

5 is a diagram illustrating an example of a flowchart of an intra prediction encoding method according to the present invention.

The intra prediction encoder 139a receives the current macro block to be predictively encoded and performs prediction encoding on the current macro block in nine intra 4x4 modes as shown in FIGS. Step 510). As a result, among each of the sixteen 4x4 subblocks included in the current macroblock, one intra 4x4 mode having the lowest cost among the nine intra 4x4 modes as shown in FIGS. Is determined.

The cost is a function value that represents the magnitude of the accuracy of prediction encoding and the amount of bits generated. Examples of functions for measuring cost include sum of absolute difference (SAD), sum of absolute transformed difference (SATD), sum of squared difference (SSD), mean of absolute difference (MAD), or Lagrange function. There is, but is not limited to this. The cost using the SAD function, for example, among the functions for measuring the cost is the sum of the absolute value of the difference between the predicted value of each pixel of the current sub-block (or macro block) and the actual pixel values.

The prediction mode determiner 139b receives the intra 4x4 cost of each sub-block determined in operation 510 from the intra prediction encoder 139a, adds them all together, calculates _{intra4x4cost,} and compares _intra4x4cost with a predetermined threshold Th _4x4 . (Step 520).

If _{intra4x4cost is} greater than or equal to a predetermined threshold Th _4x4 , the prediction mode determiner 139b skips intra 16x16 prediction and determines an intra 4x4 mode as an intra prediction mode of the current macroblock (step 580). If intra4x4cost is greater than or equal to a predetermined threshold Th _4x4 , even if intra16x16 prediction is performed, intra16x16 cost is skipped because intra16x16cost is generally higher than intra4x4cost.

However, if _intra4x4cost is smaller than the predetermined threshold Th _4x4 , the frequency Na of the prediction mode with the highest occurrence frequency among the 16 intra prediction modes determined for each subblock is greater than or equal to the predetermined threshold Th _a or determined for each subblock. Among the 16 intra prediction modes, it is determined whether the number Nb of different prediction modes is equal to or less than a predetermined threshold Th _b (operation 530).

Step 530 is a process of determining whether image data of the current macroblock is flat with a certain direction. The flatness of the image means that the change of pixel values is not large. If the image data of the current macroblock is not flat with a constant direction, it is not necessary to perform intra 16x16 prediction, and if flat, intra 16x16 prediction is performed. Meanwhile, one embodiment of the predetermined threshold Th _a is 12 (75% of the sub blocks included in the current macro block), and another embodiment of the other predetermined threshold Th _b is 3 (the sub block included in the current macro block). About 19%).

If Na is greater than or equal to the predetermined threshold Th _a or Nb is not less than or equal to the predetermined threshold Th _b , the prediction mode determiner 139b skips the intra 16x16 prediction and uses the intra 4x4 mode as the intra prediction mode of the current macroblock. Determine (step 580).

However, when Na is greater than or equal to the predetermined threshold Th _a or Nb is smaller than or equal to the predetermined threshold Th _b , the prediction mode determiner 139b controls the intra prediction encoder 139a to perform intra 16x16 prediction, and the prediction mode determiner 139b performs prediction encoding in accordance with four intra 16x16 modes (operation 540). The prediction mode determiner 139b compares the respective costs obtained as a result of predicting the four intra 16x16 modes to determine one intra 16x16 mode having the lowest cost.

The prediction mode determiner 139b receives an intra16x16cost which is a cost of the intra 16x16 mode of the current macroblock determined in operation 540 from the intra prediction encoder 139a, and compares the intra16x16cost with a predetermined threshold Th _16x16 (operation 550). ).

If intra16x16cost is smaller than the predetermined threshold Th _16x16 , the prediction mode determiner 139b determines the intra 16x16 mode as the intra prediction mode of the current macroblock (step 560). In the case of the intra 16x16 mode, the number of mode information to be transmitted is one, whereas in the case of the intra 4x4 mode, one mode must be transmitted for every 16 sub blocks, so the number of modes to be transmitted is 16. Therefore, if _intra4x4cost is smaller than the predetermined threshold Th _4x4 and intra16x16cost is smaller than the predetermined threshold Th _16x16 , the intra 16x16 mode has a little bit of information to transmit than the intra4x4 mode and thus has more gain in bitrate. The intra 16x16 mode is determined as the intra prediction mode of the current macroblock without performing.

However, if intra16x16cost is greater than or equal to the predetermined threshold Th _16x16 , the prediction mode determiner 139b compares the sizes of intra4x4cost and intra16x16cost obtained in step 520 (step 570).

If intra16x16cost is less than intra4x4cost, the intra 16x16 mode is selected as the intra prediction mode of the current macroblock (step 560). If intra16x16cost is greater than intra4x4cost, the intra 4x4 mode is selected as the intra prediction mode of the current macroblock (580). step).

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). It includes being. 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.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

 As described above, according to the present invention, the intra prediction mode of an image can be determined at a higher speed by reducing the amount of computation required to determine the intra prediction mode without degrading the intra prediction encoding performance.

Claims (14)

In the intra prediction encoding method of an image, Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And If the frequency of the prediction mode having the highest occurrence frequency among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is determined to be the final for intra prediction encoding of the input block. Determining a predictive mode. According to claim 1, After performing intra prediction encoding on the input block in the sub block unit, if the cost when intra prediction encoding in the sub block unit is equal to or greater than a predetermined threshold value Th1, the intra prediction mode for each sub block is intra prediction of the input block. Determining the prediction mode for encoding. The method of claim 2, When the cost in the case of intra prediction coding in the sub-block unit is smaller than the threshold value Th1, the frequency of the prediction mode having the highest occurrence frequency among the plurality of intra prediction modes determined for each sub block is compared with the threshold value Th _a . Method for carrying out the. According to claim 1, If the frequency of the prediction mode having the highest occurrence frequency among the plurality of intra prediction modes determined for each sub-block is greater than or equal to the threshold Th _a , intra prediction encoding is performed on the input block size basis to predict a plurality of input block size units. Among modes, further comprising determining an intra prediction mode of the input block. 5. The method of claim 4, If the cost of intra prediction coding in units of the input block size is smaller than a predetermined threshold Th2, determining the determined prediction mode in unit size of the input block as a final prediction mode for intra prediction encoding of the input block. Method comprising a. In the intra prediction encoding method of an image, Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And Among the plurality of intra prediction modes determined for each sub block, if the number of different prediction modes is larger than a predetermined threshold Th _b , the intra prediction mode for each sub block is determined as a final prediction mode for intra prediction encoding of the input block. The method comprising the step of determining. The method according to claim 6, After performing intra prediction encoding on the input block in the sub block unit, if the cost when intra prediction encoding in the sub block unit is equal to or greater than a predetermined threshold value Th1, the intra prediction mode for each sub block is intra prediction of the input block. Determining the prediction mode for encoding. 8. The method of claim 7, If the cost in the case of intra prediction coding in the sub-block unit is less than the threshold value Th1, the comparison between the number of different prediction modes and the threshold value Th _b of the plurality of intra prediction modes determined for each sub block is performed. Performing. The method according to claim 6, If the number of different prediction modes among the plurality of intra prediction modes determined for each sub-block is less than or equal to the threshold Th _b , intra prediction encoding is performed in units of the input block size to predict the plurality of input block size unit prediction modes. Among others, the method further comprises determining an intra prediction mode of the input block. The method of claim 9, If the cost of intra prediction coding in units of the input block size is smaller than a predetermined threshold Th2, determining the determined prediction mode in unit size of the input block as a final prediction mode for intra prediction encoding of the input block. Method comprising a. In the intra prediction encoding apparatus of an image, An intra prediction encoder configured to perform intra prediction encoding on an input block in units of a predetermined size, and to determine an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And If the frequency of the prediction mode having the highest occurrence frequency among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is determined to be the final for intra prediction encoding of the input block. And a prediction mode determiner configured to determine a typical prediction mode. In the intra prediction encoding apparatus of an image, An intra prediction encoder configured to perform intra prediction encoding on an input block in units of a predetermined size, and to determine an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And Among the plurality of intra prediction modes determined for each sub block, if the number of different prediction modes is larger than a predetermined threshold Th _b , the intra prediction mode for each sub block is determined as a final prediction mode for intra prediction encoding of the input block. Apparatus comprising a prediction mode determiner for determining to. Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And If the frequency of the prediction mode having the highest occurrence frequency among the plurality of intra prediction modes determined for each sub block is less than _a predetermined threshold Th _a , the intra prediction mode for each sub block is determined to be the final for intra prediction encoding of the input block. A computer-readable recording medium having recorded thereon a program for realizing an intra prediction coding method of an image, comprising the step of determining the predictive mode. Performing intra prediction encoding on an input block in units of sub blocks having a predetermined size, and determining an intra prediction mode for each sub block included in the input block among a plurality of sub block prediction modes; And Among the plurality of intra prediction modes determined for each sub block, if the number of different prediction modes is larger than a predetermined threshold Th _b , the intra prediction mode for each sub block is determined as a final prediction mode for intra prediction encoding of the input block. And a computer readable recording medium having recorded thereon a program for realizing the method of intra prediction encoding of an image.

Images