JP2013090120A - Image coding method, image decoding method, image coder, image decoder, and program thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve coding efficiency by reducing an intra prediction error by introducing adaptive reference pixel generation processing according to a coding condition in intra prediction.SOLUTION: An intra prediction block size reading unit 505 reads the size of an intra prediction block. An interpolation filter selection unit 506 sets the tap length of the interpolation filter required to generate a reference pixel, on the basis of the size of the intra prediction block. A reference pixel value generation unit 507 performs file processing using an interpolation filter of the set tap length, and generates an intra reference pixel value to use for generation of an intra prediction signal.

Description

  The present invention relates to a high-efficiency image signal encoding / decoding method, and more particularly to a technique for encoding or decoding an image using intra prediction.

  The moving image coding algorithms are roughly classified into interframe coding (inter coding) and intraframe coding (intra coding). Interframe coding is an approach for compressing information by using the correlation in the time direction in a moving image. A typical example is inter-frame prediction using motion compensation. On the other hand, intra-frame coding is an approach for compressing information using intra-frame correlation. In JPEG and MPEG-2, an approach using discrete cosine transform (DCT) is used, and in JPEG 2000, an approach using discrete wavelet transform is taken.

  H. In H.264 / AVC, prediction in the spatial direction is performed in addition to the above-described transform coding (see Non-Patent Document 1). In-screen prediction in which prediction is performed from the same screen in the spatial dimension. This intra prediction is performed on a block basis. In H.264 / MPEG-4 AVC, three types of block sizes (4 × 4, 8 × 8, 16 × 16) can be used for luminance signals. In each block size, a plurality of prediction modes can be selected. Nine types of modes are prepared for 4 × 4 and 8 × 8 block sizes, and four types of modes are prepared for 16 × 16 block sizes. For color difference signals, only an 8 × 8 block size is available, and the prediction direction is the same as for 16 × 16 blocks for luminance signals. However, the association between the mode number and the prediction direction is different.

  In these various block sizes and modes, in all cases, the pixels generated by intra prediction are the same value without changing the value of the nearest pixel to the encoding target block on the adjacent block. make a copy.

  As a specific example, FIG. 12 shows a case where the encoding target block is a 4 × 4 block of luminance signal and vertical prediction (prediction mode 0). In the following description, a luminance signal is assumed unless otherwise specified. As shown in FIG. 12 (A), X from the upper left block of the encoding target block, A, B, C, D from the upper block, E, F, G, H from the upper right block, and The pixel values indicated by I, J, K, and L from the block on the left are used for prediction. In prediction mode 0, the value of A (73) is copied to the four pixels immediately below for prediction in the vertical direction. Thereafter, similarly, the B value (79), the C value (86), and the D value (89) are respectively copied to the four pixels immediately below. As a result, the prediction pixel value of the encoding target block is as shown in FIG.

  Depending on the position where the encoding target block exists, there may be no block to be referred to. In that case, prediction is possible by substituting the value of 128 or the value of the adjacent pixel. For example, in the block in the top row of the screen, since 9 pixels from X to H cannot always be referred to, 128 is used. If the upper left and upper blocks exist but the upper right does not exist, the prediction pixel is generated by substituting the value of D into E, F, G, and H.

  H. As an approach for improving the intra prediction of H.264 / AVC, a method that subdivides eight prediction directions and supports 33 prediction directions has been proposed. This method aims to reduce the prediction error due to the coarseness of the granularity in the prediction direction.

Jun Okubo, Shinya Tsuno, Yoshihiro Kikuchi, Teruhiko Suzuki: “H.264 / AVC textbook revised third edition”, Impress, pp.110-116, 2009

  In the above-described intra prediction, the generation of reference pixels greatly affects the prediction performance. In the case of prediction in an oblique direction (prediction other than horizontal / vertical / DC prediction), the reference pixel value is the pixel value at the decimal pixel position. This pixel value is generated by interpolation using a 2-tap bilinear filter. The filter uses fixed values for the tap length and the filter coefficient regardless of the encoding condition (quantization step width, etc.). However, since the reference pixel value is a decoded pixel value located in the vicinity of the block, its characteristic changes depending on the encoding condition. For this reason, in the conventional intra prediction, the change in the characteristic of the reference pixel value according to the encoding condition is not sufficiently considered, and there is room for improvement in improving the encoding efficiency.

  The present invention has been made in view of such circumstances, and pays attention to reference pixels used for intra prediction, and introduces an adaptive reference pixel generation process according to encoding conditions, thereby reducing intra prediction errors. The purpose is to establish a highly efficient intra coding method.

  First, define terms. Hereinafter, a region where encoding is performed using intra prediction is referred to as an intra prediction block, and a reference pixel used for intra prediction is referred to as an intra reference pixel.

  In the present invention, a reference pixel value for intra prediction is generated for an intra reference pixel based on adaptive filter selection, and the intra prediction residual is reduced.

  First, an intra reference pixel existence region (hereinafter referred to as an intra reference pixel region) is identified for an intra prediction block to be encoded. The intra reference pixel is a neighboring pixel of the intra prediction block, and is determined according to the size of the intra prediction block and the intra prediction mode.

  FIG. 1 shows an example of an intra reference pixel. FIG. 1A shows an example of an intra reference pixel when the intra prediction mode is vertical prediction, and FIG. 1B shows an example of an intra reference pixel when the intra prediction mode is horizontal prediction. Show.

  In FIG. 1, a square area corresponds to a pixel. P0 represents a pixel in the encoding target block, P1 represents an encoded pixel, and P2 and P3 represent an intra reference pixel for a pixel group in the encoding target block. Thus, the reference pixel differs depending on the intra prediction mode. A region where intra reference pixels necessary for realizing all the prepared intra prediction modes exists is referred to as an intra reference pixel region. An example of the intra reference pixel region is shown in FIG.

  In the case of oblique intra prediction (prediction other than horizontal prediction / vertical prediction / DC prediction), interpolation processing is performed on pixel values in the intra reference pixel area to generate intra reference pixels. In this generation, a filter to be used for interpolation is adaptively selected based on coding parameters that affect the characteristics of the decoded image to reduce intra prediction errors.

  In this selection of the interpolation filter, the longer the intra prediction block size is, the longer the tap length interpolation filter is selected, and the smaller the intra prediction block quantization parameter is, the longer the tap length interpolation filter is selected.

  This is because the larger the size of an intra prediction block, the more complex the image of the entire block. By using a long tap length interpolation filter, it is possible to generate reference pixels with high prediction accuracy. is there. On the other hand, if the size of the intra prediction block is small, the prediction accuracy can be sufficiently maintained even with an interpolation filter having a short tap length. In addition, when the quantization parameter of the intra prediction block is small, the image often has a complex pattern, and an interpolation filter with a long tap length is appropriate for generating reference pixels with high prediction accuracy. On the other hand, when the quantization parameter of the intra prediction block is large, since the image is often flat, the prediction accuracy can be maintained even with an interpolation filter having a short tap length.

  Specifically, the present invention generates intra-prediction signals using spatial inter-pixel prediction, and intra prediction in image coding in which irreversible encoding processing is applied to a residual signal between the prediction signal and the original signal. For each reference pixel, the tap length of the interpolation filter necessary to generate the reference pixel is set and set based on either the size of the intra prediction block and / or the quantization parameter of the intra prediction block. The filter processing for generating the reference pixel is performed using the interpolation filter having the tap length, the intra prediction signal corresponding to the designated intra prediction mode is generated using the generated reference pixel, and the generated intra prediction is generated. An intra prediction error signal is generated from a difference signal between the signal and the original signal, and the intra prediction error signal is encoded as a target for irreversible encoding processing. It is characterized in.

  In addition, the present invention generates a prediction signal using spatial inter-pixel prediction, and performs image decoding using a residual signal between the prediction signal and an original signal. The prediction error signal, the intra prediction mode, and the size of the intra prediction block are decoded, the reference pixel for intra prediction is identified based on the size of the intra prediction mode and the intra prediction block, and the intra prediction block for the reference pixel for the intra prediction. The tap length of the interpolation filter necessary for generating the reference pixel is set based on one or both of the size of the block and the quantization parameter of the intra prediction block, and using the interpolation filter of the set tap length, A filter process for generating a reference pixel is performed, and a decoded intra prediction is generated using the generated reference pixel. Generates an intra prediction signal corresponding to the mode, the the generated intra prediction signal using the intra prediction error signal, and generates a decoded signal of the decoding target area.

  Further, in the present invention, when setting the tap length of the interpolation filter in the image encoding or the image decoding, a longer tap length is set when the size of the intra prediction block is larger than when the size of the intra prediction block is small. Alternatively, a longer tap length is set when the quantization step size indicated by the quantization parameter of the intra prediction block is smaller than when the quantization step size is large.

  According to the present invention, it is possible to generate an intra reference pixel value close to a prediction target pixel value. As a result, the code amount can be reduced through reduction of the intra prediction error.

It is a figure which shows the example of an intra reference pixel. It is a figure which shows one structural example of the moving image encoder to which this invention is applied. It is a figure which shows one structural example of the moving image decoding apparatus to which this invention is applied. It is a figure which shows the structural example of an intra prediction process part. It is a flowchart of an intra prediction process. It is a figure which shows the 1st structural example of a reference pixel production | generation part. It is a flowchart of an intra reference pixel production | generation process (example 1). It is a figure which shows the 2nd structural example of a reference pixel production | generation part. It is a flowchart of an intra reference pixel production | generation process (example 2). It is a figure which shows the structural example of the system in the case of implement | achieving a moving image encoder using a computer and a software program. It is a figure which shows the structural example of the system in the case of implement | achieving a moving image decoding apparatus using a computer and a software program. It is a figure which shows the example of the pixel production | generation method in the conventional intra prediction.

  The present invention relates to a technique relating to an intra prediction processing unit (101 in FIG. 2 and 202 in FIG. 3) in a video encoding device (FIG. 2) and a video decoding device (FIG. 3). The intra prediction processing unit performs a process common to the encoding device and the decoding device.

  In the following, an example of a moving picture coding apparatus and a moving picture decoding apparatus to which the present invention is applied will be shown first, and then an intra prediction processing unit which is an improved portion of the present invention will be described in detail.

[Configuration example of moving image encoding apparatus]
FIG. 2 is a diagram illustrating a configuration example of a moving image encoding apparatus to which the present invention is applied. In the moving image encoding apparatus 100, in the present embodiment, the part of the intra prediction processing unit 101 is different from the prior art, and the other part is H.264. It is the same as that of the structure of the conventional general moving image encoder used as H.264 other encoders.

  The video encoding apparatus 100 receives a video signal to be encoded, divides a frame of the input video signal into blocks, encodes each block, and outputs the bit stream as an encoded stream. For this encoding, the prediction residual signal generation unit 103 obtains a difference between the input video signal and the prediction signal output from the intra prediction processing unit 101 or the inter prediction processing unit 102, and outputs the difference as a prediction residual signal. To do. The transform processing unit 104 performs orthogonal transform such as discrete cosine transform (DCT) on the prediction residual signal and outputs a transform coefficient. The quantization processing unit 105 quantizes the transform coefficient and outputs the quantized transform coefficient. The entropy encoding processing unit 113 entropy encodes the quantized transform coefficient and outputs it as an encoded stream.

  On the other hand, the quantized transform coefficient is also input to the inverse quantization processing unit 106 where it is inversely quantized. The inverse transform processing unit 107 performs inverse orthogonal transform on the transform coefficient output from the inverse quantization processing unit 106, and outputs a prediction residual decoded signal.

  The decoded signal generation unit 108 adds the prediction residual decoded signal and the prediction signal output from the intra prediction processing unit 101 or the inter prediction processing unit 102 to generate a decoded signal of the encoded block to be encoded. This decoded signal is stored in the frame memory 109 for use as a reference image in the intra prediction processing unit 101 or the inter prediction processing unit 102. In addition, when referring in the inter prediction process part 102, in the in-loop filter process part 110, the image stored in the frame memory 109 is input, the filtering process which reduces an encoding distortion is performed, and after this filtering process Use the image as a reference image.

  Information such as the prediction mode set in the intra prediction processing unit 101 is stored in the intra prediction information storage unit 112, further entropy-coded in the entropy coding processing unit 113, and output as an encoded stream. Information such as a motion vector set in the inter prediction processing unit 102 is stored in the inter prediction information storage unit 111, further entropy-coded in the entropy coding processing unit 113, and output as an encoded stream.

[Configuration example of video decoding device]
FIG. 3 is a diagram illustrating a configuration example of a moving picture decoding apparatus to which the present invention is applied. In the moving picture decoding apparatus 200, in the present embodiment, the part of the intra prediction processing unit 202 is different from the prior art, and the other part is H.264. The configuration is the same as that of a conventional general video decoding device used as a decoder such as H.264 / AVC.

  The moving picture decoding apparatus 200 inputs and decodes the encoded stream encoded by the moving picture encoding apparatus 100 shown in FIG. 2, and outputs a video signal of a decoded image. For this decoding, the entropy decoding processing unit 201 receives an encoded stream, entropy-decodes the quantized transform coefficient of the decoding target block, and decodes information related to intra prediction and information related to inter prediction. The result of decoding information related to inter prediction is stored in the inter prediction information storage unit 209, and the result of decoding information related to intra prediction is stored in the intra prediction information storage unit 210.

  The inverse quantization processing unit 204 receives the quantized transform coefficient, inversely quantizes it, and outputs a decoded transform coefficient. The inverse transform processing unit 205 performs inverse orthogonal transform on the decoded transform coefficient, and outputs a prediction residual decoded signal. The decoded signal generation unit 206 adds the prediction residual decoded signal and the prediction signal output from the intra prediction processing unit 202 or the inter prediction processing unit 203, and generates a decoded signal of the decoding target block. This decoded signal is stored in the frame memory 207 for use as a reference image in the intra prediction processing unit 202 or the inter prediction processing unit 203. In addition, when referring in the inter prediction process part 203, the image stored in the frame memory 207 is input in the in-loop filter process part 208, the filtering process which reduces an encoding distortion is performed, and after this filtering process, Use the image as a reference image. Finally, the filtered image is output as a video signal.

[Configuration example of intra prediction processing unit]
The present embodiment is a technique related to intra prediction processing in the intra prediction processing unit 101 in FIG. 2 or the intra prediction processing unit 202 in FIG. 3.

  FIG. 4 shows a configuration example of the intra prediction processing unit. The intra prediction processing unit illustrated in FIG. 4 performs a process common to the video encoding device 100 and the video decoding device 200.

  The block position identification unit 301 identifies the position of the intra prediction block in the frame. The reference pixel generation unit 302 receives the intra prediction mode and the position of the intra prediction block in the frame, and generates an intra reference pixel for the block. The intra prediction value generation unit 303 receives the intra prediction mode and the intra reference pixel, performs prediction corresponding to the intra prediction mode, and outputs an intra prediction value.

[Intra prediction process flow]
FIG. 5 is a flowchart of intra prediction processing executed by the intra prediction processing unit shown in FIG.

  First, in step S101, the position of the intra prediction block in the frame is identified. Next, in step S102, an intra reference pixel for the block is generated with the intra prediction mode and the position of the intra prediction block in the frame as inputs. Next, in step S103, the intra prediction mode and the intra reference pixel are input, prediction corresponding to the intra prediction mode is performed, and an intra prediction value is generated and output.

[Configuration example 1 of reference pixel generation unit]
FIG. 6 illustrates a first configuration example of the reference pixel generation unit 302 in the intra prediction processing unit illustrated in FIG. The reference pixel generation unit 302 performs intra reference pixel generation processing with the following configuration.

  The decoded pixel value storage unit 501 stores a decoded pixel value necessary for generating a reference pixel. At this time, for example, H.R. The decoded pixel value may be stored after being subjected to a noise reduction filter such as a low-pass filter as in H.264 / AVC. Specifically, in FIG. 12, the value of A is not copied as it is, but processing such as (X + 2 × A + B) >> 2, (A + 2 × B + C) >> 2 (>> is a right bit shift operation) Showing). The decoded pixel value reading unit 502 receives the intra prediction mode, and reads the decoded pixel value stored in the decoded pixel value storage unit 501 according to the intra prediction mode. Whether the prediction mode determination unit 503 receives the intra prediction mode and the decoded pixel value read by the decoded pixel value reading unit 502, and whether interpolation of the decimal pixel position is necessary to generate a reference pixel used in the prediction mode If it is not necessary, a reference pixel value necessary for intra prediction is selected from the decoded pixel position. Otherwise, the process proceeds to the process of the intra prediction block size reading unit 505.

  In the present embodiment, the portions of the intra prediction block size storage unit 504, the intra prediction block size reading unit 505, and the interpolation filter selection unit 506, which are indicated by the dotted frame in FIG. 6, are different from the prior art.

  The intra prediction block size storage unit 504 stores the size of a block (intra prediction block) that is a target of intra prediction. H. In the case of H.264 / AVC, there are three types of block sizes: 4 × 4, 8 × 8, and 16 × 16. In the present embodiment, not only these sizes but also block sizes such as 32 × 32 may be targeted.

  The intra prediction block size reading unit 505 reads the size of the intra prediction block stored in the intra prediction block size storage unit 504.

  The interpolation filter selection unit 506 receives an intra prediction block size and an intra prediction mode as inputs, and selects an interpolation filter used for generating an intra reference pixel according to the size of the intra prediction block and the intra prediction mode. In particular, in the selection of this interpolation filter, an interpolation filter having a longer tap length is selected as the size of the intra prediction block is larger. For example, when the block size is 4 × 4, 8 × 8, an interpolation filter with a tap length of 2 or 4 is selected, and when the block size is 16 × 16 or 32 × 32, the tap length is 6 to 4. Select 8 or more interpolation filters.

  The reference pixel value generation unit 507 receives the intra prediction mode, the decoded pixel value read by the decoded pixel value reading unit 502, and the interpolation filter selected by the interpolation filter selection unit 506, and performs interpolation using the selected interpolation filter. By processing, reference pixel values necessary for intra prediction are generated.

  In the prior art, an interpolation filter used for generating an intra reference pixel according to an intra prediction mode by simply inputting an intra prediction mode output from the prediction mode determination unit 503 without reading the intra prediction block size. This is different from the present embodiment.

[Flow of Intra Reference Pixel Generation Process (Example 1)]
FIG. 7 is a flowchart of intra reference pixel generation processing (example 1). Hereinafter, a first example of the intra reference pixel generation process executed by the reference pixel generation unit 302 illustrated in FIG. 4 will be described in detail with reference to FIG.

  First, in step S201, the intra prediction mode is read. Next, in step S202, the intra prediction mode is input, and a decoded pixel value necessary for reference pixel generation is read. In step S203, the intra prediction mode is input, and it is determined whether or not the interpolation of the decimal pixel position is necessary to generate the reference pixel used in the prediction mode. If the interpolation is necessary, the process proceeds to step S205. Otherwise, the process proceeds to step S204.

  In step S204, the intra prediction mode and the decoded pixel value read in step S202 are input, a reference pixel value necessary for intra prediction is selected from the decoded pixel value, and is set as an intra reference pixel.

  On the other hand, if interpolation of decimal pixel positions is necessary, the size of the block (intra prediction block) that is the target of intra prediction is read in step S205. H. In the case of H.264 / AVC, there are three types of block sizes, 4 × 4, 8 × 8, and 16 × 16, but there may be block sizes larger than that.

  In step S206, the size of the intra prediction block and the intra prediction mode are input, and an interpolation filter used to generate the intra reference pixel is selected according to the size of the intra prediction block and the intra prediction mode. In this interpolation filter selection, an interpolation filter having a longer tap length is selected when the size of the intra prediction block is larger than when the size of the intra prediction block is small.

  In step S207, the intra prediction mode, the decoded pixel value read in step S202, and the interpolation filter selected in step S206 are input, and a reference pixel value necessary for intra prediction is generated by interpolation processing using the interpolation filter. .

  In FIG. 7, what is different from the prior art is steps S205 and S206 indicated by a dotted frame. In the prior art, the interpolation filter receives an intra prediction mode as an input, and simply receives an intra reference pixel according to the intra prediction mode. An interpolation filter to be used for the generation of was selected. In this embodiment, not only the intra prediction mode but also the intra prediction block size is read, and the interpolation filter used for generating the intra reference pixel is selected according to the size of the intra prediction block and the intra prediction mode. Different.

[Configuration Example 2 of Reference Pixel Generation Unit]
FIG. 8 illustrates a second configuration example of the reference pixel generation unit 302 in the intra prediction processing unit illustrated in FIG. The reference pixel generation unit 302 can also perform intra reference pixel generation processing with the configuration shown in FIG.

  In FIG. 8, the processes performed by the decoded pixel value storage unit 511, the decoded pixel value reading unit 512, the prediction mode determination unit 513, and the reference pixel value generation unit 517 are the same as those described in FIG.

  In the present embodiment, the quantization step size storage unit 514 stores a parameter (referred to as a QP parameter) indicating a quantization step size used for quantization of a block (intra prediction block) that is a target of intra prediction. .

  The quantization step size reading unit 515 reads the QP parameter stored in the quantization step size storage unit 514. The interpolation filter selection unit 516 receives the QP parameter and the intra prediction mode, and selects an interpolation filter used for generating an intra reference pixel according to the QP parameter and the intra prediction mode. In particular, in this interpolation filter selection, an interpolation filter having a longer tap length is selected as the QP parameter is smaller in accordance with predetermined correspondence information between the QP parameter and the tap length.

[Flow of Intra Reference Pixel Generation Processing (Example 2)]
FIG. 9 is a flowchart of intra reference pixel generation processing (example 2). Hereinafter, a second example of the intra reference pixel generation process executed by the reference pixel generation unit 302 illustrated in FIG. 4 will be described with reference to FIG.

  The processes performed in steps S211 to S214 and S217 shown in FIG. 9 are the same as the processes performed in steps S201 to S204 and S207 described in FIG.

  In the present embodiment, in step S215, a parameter (referred to as a QP parameter) indicating a quantization step size used for quantization of a block (intra prediction block) that is a target of intra prediction is read.

  In step S216, the QP parameter and the intra prediction mode are input, and an interpolation filter used for generating an intra reference pixel is selected according to the size of the intra prediction block and the intra prediction mode. In this interpolation filter selection, an interpolation filter having a longer tap length is selected when the QP parameter is smaller than when the QP parameter is large.

  The example of selecting the interpolation filter according to the size of the intra prediction block and the example of selecting the interpolation filter according to the quantization parameter have been described above. The tap length of the interpolation filter is set in consideration of both of them. You can also For example, when the size of the quantization parameter of the intra prediction block is the same, an interpolation filter having a longer tap length is set as the size of the intra prediction block is larger. In addition, when the size of the intra prediction block is the same, an interpolation filter having a longer tap length is set as the quantization parameter is smaller. For example, correspondence information indicating which interpolation filter of which tap length is used for each combination of the size of each intra prediction block and the quantization parameter value is stored in a table for each intra prediction mode. It is also possible to adaptively select an appropriate interpolation filter by selecting an interpolation filter based on this table.

  The above moving image encoding and decoding processes can be realized by a computer and a software program, and the program can be recorded on a computer-readable recording medium or provided through a network.

  FIG. 10 shows an example of a hardware configuration when the moving image encoding apparatus is configured by a computer and a software program. This system includes a CPU 700 that executes a program, a memory 701 such as a RAM that stores programs and data accessed by the CPU 700, and a video signal input unit 702 that inputs a video signal to be encoded from a camera or the like (disk device). And a storage unit 703 storing a video encoding program 704 that is a software program that causes the CPU 700 to execute the encoding process described in the embodiment of the present invention, The encoded stream generated by the CPU 700 executing the moving image encoding program 704 loaded in the memory 701 is output via, for example, the encoded stream output unit 705 (stores the encoded stream by the disk device or the like). May be a storage unit) In has become connected to each other.

  FIG. 11 shows an example of a hardware configuration when the moving image decoding apparatus is configured by a computer and a software program. This system includes a CPU 800 that executes a program, a memory 801 such as a RAM that stores programs and data accessed by the CPU 800, and an encoded stream that is input by an encoded stream encoded by the moving image encoding apparatus according to the present method. A program in which an input unit 802 (a storage unit that stores an encoded stream by a disk device or the like) and a moving image decoding program 804 that is a software program that causes the CPU 800 to execute the decoding processing described in the embodiment of the present invention are stored. A decoded video data output unit 805 that outputs the decoded video obtained by decoding the encoded stream by the storage device 803 and the moving picture decoding program 804 loaded into the memory 801 by the CPU 800 to a playback device or the like. (Decoded video data by disk device etc. It may also be) and a storage storing unit, but become connected to each other by a bus.

  As mentioned above, although embodiment of this invention was described with reference to drawings, the said embodiment is only the illustration of this invention, and it is clear that this invention is not limited to the said embodiment. Accordingly, additions, omissions, substitutions, and other modifications of the components may be made without departing from the spirit and technical scope of the present invention.

DESCRIPTION OF SYMBOLS 100 Moving image encoder 101,202 Intra prediction process part 200 Moving image decoder 302 Reference pixel production | generation part 501,511 Decoded pixel value memory | storage part 502,512 Decoded pixel value reading part 503,513 Prediction mode determination part 504 Intra prediction block Size storage unit 505 Intra prediction block size reading unit 506, 516 Interpolation filter selection unit 507, 517 Reference pixel value generation unit 514 Quantization step size storage unit 515 Quantization step size reading unit

Claims (8)

In an image encoding method that generates a prediction signal using spatial inter-pixel prediction and applies a lossy encoding process to a residual signal between the prediction signal and an original signal,
A step of setting a tap length of an interpolation filter necessary for generating the reference pixel based on either or both of the size of the intra prediction block and the quantization parameter of the intra prediction block with respect to the reference pixel of the intra prediction When,
Performing a filtering process for generating a reference pixel using an interpolation filter having a set tap length;
Generating an intra prediction signal corresponding to a designated intra prediction mode using the generated reference pixels;
Generating an intra prediction error signal from a difference signal between the generated intra prediction signal and the original signal;
And a step of encoding the intra prediction error signal as a target of irreversible encoding processing. The image encoding method according to claim 1,
In the step of setting the tap length of the interpolation filter,
A longer tap length is set when the size of the intra prediction block is smaller than when the size of the intra prediction block is small, or a longer tap when the quantization step size indicated by the quantization parameter of the intra prediction block is large. An image encoding method characterized by setting a length. In an image decoding method for generating a prediction signal using spatial inter-pixel prediction and performing image decoding using a residual signal between the prediction signal and an original signal,
Decoding an intra prediction error signal, an intra prediction mode, and an intra prediction block size in the input encoded stream;
Identifying a reference pixel for intra prediction based on the intra prediction mode and the size of the intra prediction block;
A step of setting a tap length of an interpolation filter necessary for generating the reference pixel based on either or both of the size of the intra prediction block and the quantization parameter of the intra prediction block with respect to the reference pixel of the intra prediction When,
Performing a filtering process for generating a reference pixel using an interpolation filter having a set tap length;
Generating an intra prediction signal corresponding to the decoded intra prediction mode using the generated reference pixel;
An image decoding method comprising: generating a decoded signal of a decoding target region using the generated intra prediction signal and the intra prediction error signal. The image decoding method according to claim 3, wherein
In the step of setting the tap length of the interpolation filter,
A longer tap length is set when the size of the intra prediction block is smaller than when the size of the intra prediction block is small, or a longer tap when the quantization step size indicated by the quantization parameter of the intra prediction block is large. An image decoding method characterized by setting a length. In an image encoding device that generates a prediction signal using spatial inter-pixel prediction and applies an irreversible encoding process to a residual signal between the prediction signal and an original signal,
Means for setting the tap length of the interpolation filter necessary for generating the reference pixel based on either or both of the size of the intra prediction block and the quantization parameter of the intra prediction block with respect to the reference pixel of the intra prediction When,
Means for performing a filter process for generating a reference pixel using an interpolation filter of a set tap length;
Means for generating an intra prediction signal corresponding to the designated intra prediction mode using the generated reference pixel;
Means for generating an intra prediction error signal from the difference signal between the generated intra prediction signal and the original signal;
An image encoding apparatus comprising: means for encoding the intra prediction error signal as a target of lossy encoding processing. In an image decoding device that generates a prediction signal using spatial inter-pixel prediction and decodes an image using a residual signal between the prediction signal and an original signal,
Means for decoding the intra prediction error signal, the intra prediction mode, and the size of the intra prediction block in the input encoded stream;
Means for identifying reference pixels for intra prediction based on the intra prediction mode and the size of the intra prediction block;
Means for setting the tap length of the interpolation filter necessary for generating the reference pixel based on either or both of the size of the intra prediction block and the quantization parameter of the intra prediction block with respect to the reference pixel of the intra prediction When,
Means for performing a filter process for generating a reference pixel using an interpolation filter of a set tap length;
Means for generating an intra prediction signal corresponding to the decoded intra prediction mode using the generated reference pixel;
An image decoding apparatus comprising: means for generating a decoded signal of a decoding target region using the generated intra prediction signal and the intra prediction error signal.   An image encoding program for causing a computer to execute the image encoding method according to claim 1.   An image decoding program for causing a computer to execute the image decoding method according to claim 3 or 4.

Images