JP2020137119A - Intra prediction device, image coding device, image decoding device, and program - Google Patents

Abstract

To further improve a prediction accuracy of an intra prediction by a prediction image composition method.SOLUTION: An intra prediction device performs an intra prediction for an image block obtained by dividing an original image. The intra prediction device comprises: a first prediction image generation part that predicts the image block from a first intra prediction mode as a directional property prediction; a second prediction image generation part that generates a second prediction image by predicting the image block by a second intra prediction mode as a non-directional property prediction; a weighting coefficient determination part that determines a weighting coefficient used when performing a weighting composition of the first prediction image and the second prediction image on the basis of at least one of the block size and a block form of the image block; and an image composition part that performs the weighting composition of the first and second prediction images by using the weighting coefficient determined by the weighting coefficient determination part.SELECTED DRAWING: Figure 3

Description

The present invention relates to an intra prediction device, an image coding device, an image decoding device, and a program.

Research on video coding methods is being conducted to compress the amount of data during transmission and storage of still images and moving images. In recent years, in video coding technology, ultra-high resolution video represented by 8K-SHV has become widespread, and AVC / H.A. is used as a method for transmitting a moving image of a huge amount of data.
264 and HEVC / H. Coding methods such as 265 are known.

Intra-prediction using spatial correlation within a frame is used in VVC evaluation software (VTM), which is a next-generation video coding method jointly standardized by MPEG and ITU (Non-Patent Documents). 1). Using the decoded reference pixels around the image block to be encoded, Planar prediction, DC prediction, and 65 directional predictions, a total of 67
From the street prediction mode, the optimum mode is selected on the encoder side, and that information is sent to the decoder side.

As a method for improving the prediction accuracy of intra-prediction, a prediction image synthesis method has been proposed in which a prediction image is generated in each of the two intra prediction modes and each pixel of the two prediction images is added to generate a new prediction image. (See Non-Patent Document 2). Specifically, 6 above
A prediction image is generated by adding a direction prediction image, which is a prediction image generated by any of the five modes of direction prediction, and a Planar prediction image, which is a prediction image generated by the Planar mode.

JVET-L1001 "Versatile Video Coding (Draft 3)" JVET-M0458 “Non-CE3: Combined-Hypothesis Intra-Prescription”

By the way, the directional prediction has a drawback that the prediction accuracy is high for a pixel at a position close to the reference pixel, but the prediction accuracy may decrease as the distance from the reference pixel increases.

However, the predictive image synthesis method described in Non-Patent Document 2 includes a directional predictive image and a Plan.
It merely averages the ar prediction image on a pixel-by-pixel basis, does not consider the drawbacks of directional prediction, and there is room for improvement in further improving the prediction accuracy of intra prediction.

Therefore, an object of the present invention is to provide an intra-prediction device, an image coding device, an image decoding device, and a program that further enhance the prediction accuracy of intra-prediction by a predictive image synthesis method.

The intra prediction device according to the first aspect performs intra prediction for an image block obtained by dividing an original image. The intra prediction device includes a first prediction image generation unit that predicts the image block by a first intra prediction mode that is directional prediction and generates a first prediction image, and a second intra prediction mode that is non-directional prediction. Based on at least one of the block size and block shape of the image block and the second prediction image generation unit that predicts the image block and generates the second prediction image, the first prediction image and the second prediction image. The first predicted image and the second predicted image are weighted and synthesized using the weighting coefficient determining unit that determines the weighting coefficient used when weighting and synthesizing the images and the weighting coefficient determined by the weighting coefficient determining unit. It is equipped with an image compositing unit.

The image coding device according to the second aspect includes an intra prediction device according to the first aspect.

The image decoding device according to the third aspect includes an intra prediction device according to the first aspect.

The program according to the fourth aspect causes the computer to function as the intra prediction device according to the first aspect.

According to the present invention, it is possible to provide an intra-prediction device, an image coding device, an image decoding device, and a program that further enhance the prediction accuracy of intra-prediction by a predictive image synthesis method.

It is a figure which shows the structure of the image coding apparatus which concerns on embodiment. It is a figure which shows the prediction mode of the intra prediction which concerns on embodiment. It is a figure which shows the structure of the intra prediction part of the image coding apparatus which concerns on embodiment. It is a figure which shows the operation example of the intra prediction part which concerns on embodiment. It is a figure which shows the operation example 1 of the weighting coefficient determination part which concerns on embodiment. It is a figure which shows the operation example 2 of the weighting coefficient determination part which concerns on embodiment. It is a figure which shows the structure of the image decoding apparatus which concerns on embodiment. It is a figure which shows the structure of the intra prediction part of the image decoding apparatus which concerns on embodiment. It is a figure which shows the operation flow example of the intra prediction part which concerns on embodiment.

The image coding apparatus and the image decoding apparatus according to the embodiment will be described with reference to the drawings.
The image coding device and the image decoding device according to the embodiment encode and decode a moving image represented by MPEG, respectively. In the description of the drawings below, the same or similar parts are designated by the same or similar reference numerals.

<Configuration of image coding device>
First, the image coding apparatus according to the present embodiment will be described. FIG. 1 is a diagram showing a configuration of an image coding device 1 according to the present embodiment.

As shown in FIG. 1, the image coding apparatus 1 includes a block dividing unit 100, a subtracting unit 110, and the same.
It has a conversion / quantization unit 120, an entropy coding unit 130, an inverse quantization / inverse conversion unit 140, a synthesis unit 150, a memory 160, and a prediction unit 170.

The block division unit 100 divides the original image, which is an input image for each frame (or picture) constituting the moving image, into a plurality of image blocks, and the image block obtained by the division is subtracted from the subtraction unit 1.
Output to 10. The size of the image block is, for example, 32 × 32 pixels, 16 × 16 pixels, 8
It is × 8 pixels, 4 × 4 pixels, or the like. The shape of the image block is not limited to a square and may be a rectangle (rectangle). The image block is a unit for encoding by the image coding device 1 (encoding target block) and a unit for decoding by the image decoding device (decoding target block). Such an image block may be called a CU (Coding Unit).

The subtraction unit 110 calculates a prediction residual representing a difference (error) between the coding target block input from the block division unit 100 and the prediction image obtained by predicting the coding target block by the prediction unit 170. Specifically, the subtraction unit 110 calculates the prediction residual by subtracting each pixel value of the prediction image from each pixel value of the block, and outputs the calculated prediction residual to the conversion / quantization unit 120.

The conversion / quantization unit 120 performs orthogonal conversion processing and quantization processing in block units. conversion·
The quantization unit 120 includes a conversion unit 121 and a quantization unit 122.

The conversion unit 121 performs orthogonal conversion processing on the predicted residual input from the subtraction unit 110 to calculate the orthogonal conversion coefficient, and outputs the calculated orthogonal conversion coefficient to the quantization unit 122. The orthogonal transform is, for example, the Discrete Cosine Transform (DCT).
form) and discrete sine conversion (DST: Discrete Sine Transfer)
m), Karhunen-Loeve Transfer (KLT: Karhunen-Loeve Transfo)
rm) etc.

The quantization unit 122 uses the orthogonal conversion coefficient input from the conversion unit 121 as a quantization parameter (Q).
It is quantized using p) and the quantization matrix, and the quantized orthogonal conversion coefficient is output to the entropy coding unit 130 and the inverse quantization / inverse conversion unit 140. The quantization parameter (Qp) is
It is a parameter that is commonly applied to each orthogonal conversion coefficient in the block and determines the roughness of quantization. The quantization matrix is a matrix having a quantization value as an element when quantizing each orthogonal conversion coefficient.

The entropy coding unit 130 performs entropy coding on the orthogonal conversion coefficient input from the quantization unit 122, performs data compression to generate coded data (bit stream), and image-codes the coded data. Output to the outside of device 1. For entropy coding, Huffman code or CABAC (Context-based Adaptive Binar)
y Arithmetic Coding; context-adaptive binary arithmetic coding) and the like can be used. The entropy coding unit 130 receives information such as syntax related to prediction from the prediction unit 170, and also performs entropy coding of the input information.

The inverse quantization / inverse conversion unit 140 performs inverse quantization processing and inverse orthogonal conversion processing in block units. The inverse quantization / inverse conversion unit 140 includes an inverse quantization unit 141 and an inverse conversion unit 142.

The inverse quantization unit 141 performs an inverse quantization process corresponding to the quantization process performed by the quantization unit 122. Specifically, the inverse quantization unit 141 restores the orthogonal conversion coefficient by inversely quantizing the orthogonal conversion coefficient input from the quantization unit 122 using the quantization parameter (Qp) and the quantization matrix. , The restored orthogonal conversion coefficient is output to the inverse conversion unit 142.

The inverse conversion unit 142 performs an inverse orthogonal conversion process corresponding to the orthogonal conversion process performed by the conversion unit 121. For example, when the conversion unit 121 performs the discrete cosine transform, the inverse transform unit 142 performs the inverse discrete cosine transform. The inverse conversion unit 142 performs an inverse orthogonal conversion process on the orthogonal conversion coefficient input from the inverse quantization unit 141 to restore the predicted residual, and the restored predicted residual, which is the restored predicted residual, is combined with the composite unit 150. Output to.

The synthesizing unit 150 synthesizes the restoration prediction residual input from the inverse conversion unit 142 with the prediction image input from the prediction unit 170 in pixel units. The synthesizing unit 150 adds each pixel value of the restoration prediction residual and each pixel value of the predicted image to reconstruct (decode) the coded block, and outputs the decoded image of the decoded block unit to the memory 160. Such a decoded image may be referred to as a reconstructed image.

The memory 160 stores the decoded image input from the compositing unit 150. Memory 160
The decoded image is stored in frame units. The memory 160 predicts the stored decoded image in the prediction unit 1.
Output to 70. A loop filter may be provided between the compositing unit 150 and the memory 160. A part of the memory 160 may be included in the prediction unit 170.

The prediction unit 170 makes a prediction in block units. The prediction unit 170 is an inter prediction unit 171.
And an intra prediction unit 172 and a switching unit 173.

The inter-prediction unit 171 uses the decoded image stored in the memory 160 as a reference image, calculates a motion vector by a method such as block matching, predicts a block to be encoded, and generates an inter-prediction image. The inter-prediction image is output to the switching unit 173.

The inter-prediction unit 171 selects the optimum inter-prediction method from inter-prediction using a plurality of reference images (typically bi-prediction) and inter-prediction using one reference image (one-way prediction). Inter-prediction is performed using the selected inter-prediction method. The inter-prediction unit 171 inputs information (motion vector, etc.) related to the inter-prediction to the entropy coding unit 130.
Output to.

The intra prediction unit 172 generates an intra prediction image by referring to the decoded pixel values around the coded block among the decoded images stored in the memory 160, and outputs the generated intra prediction image to the switching unit 173. To do. Further, the intra prediction unit 172 outputs the syntax related to the selected prediction mode to the entropy coding unit 130. In the following, a block that is the target of intra-prediction will be referred to as an intra-prediction target block.

The intra prediction unit 172 selects the optimum prediction mode to be applied to the intra prediction target block from the plurality of prediction modes, and predicts the intra prediction target block using the selected prediction mode.

FIG. 2 is a diagram showing a prediction mode of intra prediction according to the present embodiment. As shown in FIG. 2, there are 67 prediction modes from 0 to 66. Prediction mode mode "0" is Pla
nar prediction, prediction mode mode "1" is DC prediction, prediction mode mode "1"
"2" to "66" are direction predictions. In the directionality prediction, the direction of the arrow indicates the prediction direction, the starting point of the arrow indicates the position of the pixel to be predicted, and the end point of the arrow indicates the position of the reference pixel used for predicting the pixel to be predicted. Modes "2" to "33" are prediction modes that mainly refer to the reference pixel on the left side of the target block for intra prediction. On the other hand, modes "35" to "6"
6 ”is a prediction mode in which the reference pixel on the upper side of the target block for intra prediction is mainly referred to.

The switching unit 173 switches between the inter prediction image input from the inter prediction unit 171 and the intra prediction image input from the intra prediction unit 172, and outputs one of the prediction images to the subtraction unit 110 and the synthesis unit 150.

FIG. 3 is a diagram showing the configuration of the intra prediction unit 172 according to the present embodiment. The intra prediction unit 172 corresponds to an intra prediction device provided in the image coding device 1.

As shown in FIG. 3, the intra prediction unit 172 includes a memory 160a, a first prediction image generation unit 172a, a second prediction image generation unit 172b, a weighting coefficient determination unit 172c, and an image composition unit 1.
It has 72d and.

The memory 160a is a part of the memory 160 shown in FIG. The memory 160a stores reference pixels, which are decoded pixels that are referred to during intra-prediction.

The first prediction image generation unit 172a predicts the intra prediction target block by the first intra prediction mode, which is the direction prediction, generates a direction prediction image (first prediction image), and images the generated direction prediction image. It is output to the synthesis unit 172d. Specifically, the first prediction image generation unit 172
a refers to the reference pixel stored in the memory 160a, and generates a directional prediction image in any mode of 65 directional predictions.

The second prediction image generation unit 172b predicts the intra prediction target block by the second intra prediction mode, which is non-directional prediction, generates a prediction image (second prediction image), and generates a prediction image (second prediction image), and the generated prediction image is generated by the image synthesis unit 172d. Output to. Specifically, the second prediction image generation unit 172b refers to the reference pixel stored in the memory 160a and generates a directional prediction image in a predetermined non-directional second intra prediction mode.

The second intra prediction mode may be any prediction mode as long as it is a non-directional prediction mode, but an example in which the second intra prediction mode is a Planar prediction will be described in the present embodiment.

The weighting coefficient determining unit 172c determines the weighting coefficient based on at least one of the block size and the block shape of the intra-prediction target block, and outputs the determined weighting coefficient to the image composition unit 172d. The weighting coefficient output by the weighting coefficient determining unit 172c is a directional prediction image generated by the first prediction image generation unit 172a and a Pla generated by the second prediction image generation unit 172b.
It is used when weighting and synthesizing nar prediction images.

The image synthesizing unit 172d uses the weighting coefficient input from the weighting coefficient determining unit 172c to generate the directional prediction image and the second predicted image generation unit 172 input from the first prediction image generation unit 172a.
The Planar prediction image input from b is weighted and synthesized, and the predicted image after weighting and synthesis is output as an intra prediction image.

FIG. 4 is a diagram showing an operation example of the intra prediction unit 172 according to the present embodiment. In FIG. 4, an example in which the intra-prediction target block has a square shape of 8 × 8 pixels is illustrated, but the intra-prediction target block does not have to be a square. The black circles in FIG. 4 represent reference pixels.

As shown in FIG. 4A, the second prediction image generation unit 172b predicts the intra-prediction target block by the Planar prediction and generates the Planar prediction image. Specifically, Pl
In the anar prediction, the predicted pixel value is generated by interpolation prediction using the four reference pixels shown in FIG. 4A.

As shown in FIG. 4B, the first prediction image generation unit 172a predicts the intra-prediction target block in the direction prediction mode and generates a direction prediction image. In directional prediction, the reference pixel is extrapolated along the prediction direction to generate the predicted pixel value. Therefore, although the prediction accuracy is high for the pixel at the position close to the reference pixel, the prediction is made as the distance from the reference pixel increases. It has the disadvantage that the accuracy can be low.

The weighting coefficient determining unit 172c determines the weighting coefficient based on at least one of the block size and the block shape of the intra-prediction target block. In the present embodiment, the weighting coefficient determining unit 172c has a weighting coefficient α (second weighting coefficient) applied to the Planar prediction image and
The weighting coefficient β (first weighting coefficient) applied to the direction prediction image is determined.

As shown in FIG. 4C, the image synthesizing unit 172d applies a weighting coefficient β to each pixel of the directional prediction image generated by the first prediction image generation unit 172a, and the second prediction image generation unit 1
A weighting coefficient α is applied to each pixel of the Planar prediction image generated by 72b, the direction prediction image and the Planar prediction image are combined in pixel units, and the combined prediction image is output as an intra prediction image.

FIG. 5 is a diagram showing an operation example 1 of the weighting coefficient determining unit 172c according to the present embodiment. In FIG. 5, the “small side length” represents the length of the short side of the rectangular intra-prediction target block in terms of the number of pixels. That is, the "small side length" represents the block size of the intra-prediction target block. On the other hand, “α: β” in FIG. 5 represents the ratio of the weighting coefficient α to the weighting coefficient β.

Here, "small side length" is presented as an example showing the block size of the intra-prediction target block, but "block width + block height" and "block width x block height" are other examples. Etc. may be used.

As shown in FIG. 5, the weighting coefficient determining unit 172c increases the ratio of the weight (α) of the Planar prediction image to the weight (β) of the direction prediction image as the block size of the intra prediction target block increases. In addition, the weighting coefficients α and β are determined.

In the example shown in FIG. 5, when the block size is “8”, the weight (α) of the Planar predicted image and the weight (β) of the directional predicted image are equal. In this case, the image synthesizing unit 172d outputs a predicted image obtained by averaging the Planar predicted image and the directional predicted image as an intra predicted image. In the following, the block size "8" will be referred to as a reference block size.

When the block size is "4", which is smaller than the reference block size, the weighting coefficient determination unit 1
In 72c, the weighting coefficients α and β are determined so that the weight (α = 1) of the Planar predicted image is smaller than the weight (β = 2) of the directional prediction image. In this case, the ratio of the Planar predicted image to the intra predicted image output by the image synthesizing unit 172d becomes small.

On the other hand, the weighting coefficient determining unit 172c has weighting coefficients α and β so that the weight (α) of the Planar predicted image becomes larger than the weight (β) of the directional prediction image as the block size becomes larger than the reference block size. To determine. In this case, the ratio of the Planar predicted image to the intra predicted image output by the image synthesizing unit 172d becomes large.

Here, in the directionality prediction, the larger the block size of the intra-prediction target block, the lower the prediction accuracy of the pixel far from the reference pixel. Therefore, the prediction accuracy can be improved by determining the weighting coefficients α and β so as to increase the proportion of the Planar prediction image and decrease the proportion of the directional prediction image.

The method for determining the weighting coefficients α and β shown in FIG. 5 is an example, and the specific values of the weighting coefficients α and β are not limited to the example shown in FIG.

FIG. 6 is a diagram showing an operation example 2 of the weighting coefficient determining unit 172c according to the present embodiment. In FIG. 6, "the larger of might / height or height / width" is
For the intra-prediction target block, it is the larger value of the width (wise) / height (height) value and the height (height) / width (wise) value. That is, "
“Which / height or height / width is larger” is a value that becomes larger as the intra-prediction target block has an elongated shape. On the other hand, "α: β" in FIG. 6 is
It represents the ratio of the weighting coefficient α to the weighting coefficient β.

As shown in FIG. 6, the weighting coefficient determining unit 172c has a Pl of the weight (β) of the directional prediction image as the ratio of the long side to the short side of the rectangular intra-prediction target block increases.
The weighting coefficients α and β are determined so as to reduce the ratio of the weights (α) of the anar predicted image.
The short side means the short side of the rectangle, and the long side means the long side of the rectangle.

In other words, the weighting coefficient determination unit 172c reduces the ratio of the weight (α) of the Planar prediction image to the weight (β) of the direction prediction image as the intra-prediction target block has an elongated shape. And β are determined.

In the example shown in FIG. 6, "wise / height or height / width"
When "the larger one" is "1", that is, when the intra-prediction target block is square, the weight (α) of the Planar prediction image and the weight (β) of the direction prediction image are equal. In this case, the image synthesizing unit 172d outputs a predicted image obtained by averaging the Planar predicted image and the directional predicted image as an intra predicted image. In the following, the square intra-prediction target block will be referred to as a reference shape.

In the weight coefficient determination unit 172c, as the intra-prediction target block changes from the reference shape (square) to the elongated shape, the weight (α) of the Planar prediction image becomes the weight of the direction prediction image (
The weighting coefficients α and β are determined so as to be smaller than β). In this case, the image composition unit 172
The ratio of the Planar predicted image to the intra predicted image output by d becomes smaller.

Here, in the directionality prediction, the longer the intra-prediction target block is, the less likely it is to predict a pixel far away from the reference pixel, and the higher the prediction accuracy can be. Therefore, the prediction accuracy can be improved by determining the weighting coefficients α and β so as to reduce the proportion of the Planar prediction image and increase the proportion of the directional prediction image.

The method for determining the weighting coefficients α and β shown in FIG. 6 is an example, and the specific values of the weighting coefficients α and β are not limited to the example shown in FIG.

Further, the operation example 1 and the operation example 2 of the weighting coefficient determining unit 172c may be combined and carried out. For example, a table as shown in FIG. 6 may be defined separately for each value of "small side length" (that is, block size). Alternatively, "wise / height or hei"
A table as shown in FIG. 5 may be defined separately for each value of “ght / width larger”.

<Configuration of image decoding device>
Next, the image decoding device according to the present embodiment will be described. FIG. 7 is a diagram showing the configuration of the image decoding device 2 according to the present embodiment.

As shown in FIG. 7, the image decoding device 2 includes an entropy decoding unit 200, an inverse quantization / inverse conversion unit 210, a synthesis unit 220, a memory 230, and a prediction unit 240.

The entropy decoding unit 200 decodes the coded data generated by the image coding device 1 and outputs the quantized orthogonal conversion coefficient to the inverse quantization / inverse conversion unit 210. Further, the entropy decoding unit 200 acquires the syntax related to the prediction (intra-prediction and inter-prediction), and outputs the acquired syntax to the prediction unit 240.

The inverse quantization / inverse conversion unit 210 performs inverse quantization processing and inverse orthogonal conversion processing in block units. The inverse quantization / inverse conversion unit 210 includes an inverse quantization unit 211 and an inverse conversion unit 212.

The inverse quantization unit 211 performs the inverse quantization process corresponding to the quantization process performed by the quantization unit 122 of the image coding apparatus 1. The inverse quantization unit 211 dequantizes the quantization orthogonal conversion coefficient input from the entropy decoding unit 200 using the quantization parameter (Qp) and the quantization matrix to obtain the orthogonal conversion coefficient of the block to be decoded. It is restored and the restored orthogonal conversion coefficient is output to the inverse conversion unit 212.

The inverse conversion unit 212 performs an inverse orthogonal conversion process corresponding to the orthogonal conversion process performed by the conversion unit 121 of the image coding apparatus 1. The inverse conversion unit 212 performs an inverse orthogonal conversion process on the orthogonal conversion coefficient input from the inverse quantization unit 211 to restore the predicted residual, and the restored predicted residual (restored predicted residual).
Is output to the synthesis unit 220.

The synthesis unit 220 reconstructs (decodes) the original block by synthesizing the prediction residual input from the inverse conversion unit 212 and the prediction image input from the prediction unit 240 in pixel units, and blocks units. The decoded image of is output to the memory 230.

The memory 230 stores the decoded image input from the compositing unit 220. Memory 230
The decoded image is stored in frame units. The memory 230 outputs a frame-by-frame decoded image to the outside of the image decoding device 2. A loop filter may be provided between the compositing unit 220 and the memory 230. Further, a part of the memory 230 may be included in the prediction unit 240.

The prediction unit 240 makes a prediction in block units. The prediction unit 240 is an inter prediction unit 241.
And an intra prediction unit 242 and a switching unit 243.

The inter-prediction unit 241 uses the decoded image stored in the memory 230 as a reference image to predict the decoding target block by inter-prediction. The inter-prediction unit 241 generates an inter-prediction image by performing inter-prediction according to the syntax, motion vector, etc. input from the entropy decoding unit 200, and the generated inter-prediction image is switched to the switching unit 243.
Output to.

The intra prediction unit 242 refers to the decoded image stored in the memory 230, and generates an intra prediction image by predicting the decoding target block by the intra prediction based on the syntax input from the entropy decoding unit 200. The generated intra prediction image is output to the switching unit 243.

The switching unit 243 switches between the inter prediction image input from the inter prediction unit 241 and the intra prediction image input from the intra prediction unit 242, and outputs one of the prediction images to the synthesis unit 220.

FIG. 8 is a diagram showing the configuration of the intra prediction unit 242 according to the present embodiment. The intra prediction unit 242 corresponds to an intra prediction device provided in the image decoding device 2. The intra prediction unit 242 operates in the same manner as the intra prediction unit 172 provided in the image coding device 1.

As shown in FIG. 8, the intra prediction unit 242 includes a memory 230a, a first prediction image generation unit 242a, a second prediction image generation unit 242b, a weight coefficient determination unit 242c, and an image composition unit 2.
It has 42d and.

The memory 230a is a part of the memory 230 shown in FIG. 7. The memory 230a stores reference pixels, which are decoded pixels that are referred to during intra-prediction.

The first prediction image generation unit 242a predicts the intra prediction target block by the first intra prediction mode, which is the direction prediction, generates a direction prediction image (first prediction image), and images the generated direction prediction image. It is output to the synthesis unit 242d. Specifically, the first prediction image generation unit 242
a refers to the reference pixel stored in the memory 230a, and generates a directional prediction image according to the directional prediction mode indicated by the syntax input from the entropy decoding unit 200.

The second prediction image generation unit 242b predicts the intra prediction target block by the second intra prediction mode, which is a non-directional prediction, and generates a prediction image (second prediction image), and the generated prediction image is generated by the image synthesis unit 242d. Output to. Specifically, the second prediction image generation unit 242b refers to the reference pixel stored in the memory 230a and generates a directional prediction image in a predetermined non-directional second intra prediction mode. In this embodiment, the second intra prediction mode is Planar prediction.

The weighting coefficient determining unit 242c determines the weighting coefficient based on at least one of the block size and the block shape of the intra-prediction target block, and outputs the determined weighting coefficient to the image combining unit 242d. The block size and block shape of the intra-prediction target block are
It is input from the entropy decoding unit 200 to the weighting coefficient determining unit 242c. Weight coefficient determination unit 2
The operation of 42c is the same operation as the above-described operation examples 1 and 2.

The image composition unit 242d uses the weighting coefficient input from the weighting coefficient determining unit 242c to generate the directional prediction image and the second prediction image generation unit 242 input from the first prediction image generation unit 242a.
The Planar prediction image input from b is weighted and synthesized, and the predicted image after weighting and synthesis is output as an intra prediction image.

<Example of intra-prediction operation flow>
Next, an example of the operation flow of the intra prediction according to the present embodiment will be described. The intra-prediction operation is the same in the image coding device 1 and the image decoding device 2, but here, the image decoding device 2
The operation of the intra prediction (intra prediction unit 242) in the above will be described. FIG. 9 is a diagram showing an example of an operation flow of the intra prediction unit 242.

First, the entropy decoding unit 200 decodes the syntax indicating the intra prediction mode selected by the image coding device 1. When this syntax indicates a mode other than the directional prediction mode (DC prediction or Planar prediction) (step S1: NO), the intra prediction unit 242 performs the same intra prediction as before (step S7).

Second, the entropy decoding unit 200 decodes the syntax indicating whether or not to apply the predictive image synthesis method. When this syntax indicates that the prediction image composition method is not applied (step S2: NO), the intra prediction unit 242 performs the same intra prediction as before (step S7). However, when the intra prediction mode selected by the image coding apparatus 1 is the directional prediction mode, the prediction image composition method may always be applied. In this case, syntax signaling indicating whether or not to apply the predictive image synthesis method is unnecessary.

Third, the intra prediction mode selected by the image encoding device 1 is directional prediction (direction prediction).
Step S1: YES) and when applying the predictive image composition method (Step S2: YE)
S), the weighting coefficient determining unit 242c determines the weighting coefficients α and β based on at least one of the block size and the block shape of the intra-prediction target block (step S3).
..

Fourth, the first prediction image generation unit 242a predicts the intra-prediction target block by the direction prediction mode selected by the image coding device 1 and generates a direction prediction image (step S4). In addition, the second prediction image generation unit 242b predicts the intra-prediction target block by the Planar prediction and generates the Planar prediction image (step S5).

Fifth, the image synthesizing unit 242d weight-combines the directional prediction image generated in step S4 and the Planar prediction image generated in step S5 by using the weighting coefficients α and β determined in step S3. , The predicted image after weighting synthesis is output as an intra prediction image (step S6).

<Summary of Embodiment>
The image coding device 1 and the image decoding device 2 according to the present embodiment determine the weighting coefficient based on at least one of the block size and the block shape of the intra-prediction target block.
The directional prediction image and the Planar prediction image are weighted and combined using the determined weighting coefficient. As a result, in the prediction image synthesis method, the prediction accuracy of the intra prediction can be improved as compared with the case where the direction prediction image and the Planar prediction image are simply averaged.

<Other Embodiments>
In the above-described embodiment, an example in which the second intra-prediction mode is Planar prediction has been mainly described. However, the second intra prediction mode may be Plane prediction. When the second intra-prediction mode is Plane prediction, "Planar prediction" in the above-described embodiment may be read as "Plane prediction".

Alternatively, the second intra prediction mode may be DC prediction. When the second intra prediction mode is DC prediction, "Planar prediction" in the above-described embodiment may be read as "DC prediction".

A program may be provided that causes a computer to execute each process performed by the image coding apparatus 1. A program that causes a computer to execute each process performed by the image decoding device 2 may be provided. The program may be recorded on a computer-readable medium. Computer-readable media can be used to install programs on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, but may be, for example, a recording medium such as a CD-ROM or a DVD-ROM.

A circuit that executes each process performed by the image coding device 1 may be integrated, and the image coding device 1 may be configured by a semiconductor integrated circuit (chipset, SoC). A circuit that executes each process performed by the image decoding device 2 is integrated, and the image decoding device 2 is a semiconductor integrated circuit (chipset, SoC).
It may be configured by.

Although the embodiments have been described in detail with reference to the drawings, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the gist.

1: Image coding device 2: Image decoding device 100: Block division unit 110: Subtraction unit 120: Conversion / quantization unit 121: Conversion unit 122: Quantization unit 130: Entropy coding unit 140: Inverse quantization / inverse conversion Unit 141: Inverse quantization unit 142: Inverse conversion unit 150: Synthesis unit 160: Memory 160a: Memory 170: Prediction unit 171: Inter-prediction unit 172: Intra-prediction unit 172a: First prediction image generation unit 172b: Second prediction image Generation unit 172c: Weight coefficient determination unit 172d: Image composition unit 173: Switching unit 200: Entropy decoding unit 210: Inverse quantization / inverse conversion unit 211: Inverse quantization unit 212: Inverse conversion unit 220: Synthesis unit 230: Memory 230a : Memory 240: Prediction unit 241: Inter prediction unit 242: Intra prediction unit 242a: First prediction image generation unit 242b: Second prediction image generation unit 242c: Weight coefficient determination unit 242d: Image composition unit 243: Switching unit

Claims (8)

It is an intra-prediction device that makes intra-prediction for image blocks obtained by dividing the original image.
A first prediction image generation unit that predicts the image block and generates a first prediction image in the first intra prediction mode, which is a direction prediction.
A second prediction image generation unit that predicts the image block and generates a second prediction image in the second intra prediction mode, which is non-directional prediction.
A weighting coefficient determining unit that determines a weighting coefficient used when weighting and synthesizing the first predicted image and the second predicted image based on at least one of the block size and the block shape of the image block.
An intra-prediction apparatus comprising: an image synthesizing unit for weighting and synthesizing the first prediction image and the second prediction image using the weighting coefficient determined by the weighting coefficient determining unit. The intra-prediction device according to claim 1, wherein the second intra-prediction mode is Planar prediction, Plane prediction, or DC prediction. The weighting coefficient determining unit determines the weighting coefficient so that the ratio of the weight of the second predicted image to the weight of the first predicted image increases as the block size of the image block increases. The intra prediction device according to claim 1 or 2. The weighting coefficient determining unit reduces the ratio of the weight of the second predicted image to the weight of the first predicted image as the ratio of the long side to the short side of the rectangular image block increases. The intra prediction device according to any one of claims 1 to 3, wherein the weighting coefficient is determined. The weighting coefficient determining unit determines the first weighting coefficient applied to the first predicted image and the second weighting coefficient applied to the second predicted image as the weighting coefficient.
The image synthesizing unit applies the first weighting coefficient to each pixel of the first predicted image, applies the second weighting coefficient to each pixel of the second predicted image, and applies the second weighting coefficient to each pixel of the second predicted image. The intra-prediction device according to any one of claims 1 to 4, wherein the second prediction image is synthesized in pixel units. An image coding device including the intra prediction device according to any one of claims 1 to 5. An image decoding device comprising the intra prediction device according to any one of claims 1 to 5. A program for operating a computer as an intra prediction device according to any one of claims 1 to 5.

Images