JP7589791B2 - Image encoding device, image encoding method, image encoding program, image decoding device, image decoding method, image decoding program, storage method, transmission method - Google Patents

Description

The present invention relates to a technique for dividing an image into blocks and encoding and decoding the divided blocks.

In image coding and decoding, an image is divided into blocks, which are sets of a predetermined number of pixels, and coding and decoding are performed on a block-by-block basis. By dividing an image into blocks appropriately, the efficiency of intra-frame prediction, inter-frame prediction, orthogonal transformation, entropy coding, and the like is improved, and as a result, coding efficiency is improved.

Special Publication No. 2015-526008

If the blocks are not divided into blocks of appropriate size and shape, the coding efficiency decreases, and the amount of processing required in the subsequent coding and decoding increases.

The present invention has been made in view of the above circumstances, and has as its object to provide a technique for improving coding efficiency by performing block division suitable for image coding and decoding.

In order to solve the above problem, an image coding device of one embodiment of the present invention is an image coding device that divides an image into blocks and codes the divided blocks in units of the divided blocks, and includes a block division unit that recursively divides the image into rectangles of a predetermined size to generate a block to be coded, an intra prediction unit that generates a predicted image signal using image signals of coded blocks adjacent to the block to be coded, and a coding unit that codes block division information of the block to be coded, wherein the block division unit includes a quartering unit that divides the target block in the recursive division into four horizontally and vertically to generate four blocks, and a bisection unit that divides the target block in the recursive division into two horizontally or vertically to generate two blocks, and the bisection unit prohibits dividing the target block for the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division if the previous recursive division was a bisection.

Yet another aspect of the present invention is an image coding method, which divides an image into blocks and codes the divided blocks, and includes a block division step of recursively dividing the image into rectangles of a predetermined size to generate a block to be coded, an intra prediction step of generating a predicted image signal using image signals of coded blocks adjacent to the block to be coded, and a coding step of coding block division information of the block to be coded, the block division step including a quartering step of dividing the block to be coded in the recursive division into four horizontally and vertically to generate four blocks, and a bidivision step of dividing the block to be coded in the recursive division into two horizontally or vertically to generate two blocks, and the bidivision step prohibits dividing the block to be coded in the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division if the previous recursive division was bidivision.

Yet another embodiment of the present invention is an image encoding program, comprising: dividing an image into blocks;
An image coding program for coding divided blocks, the program causing a computer to execute a block division step of recursively dividing the image into rectangles of a predetermined size to generate a block to be coded, an intra prediction step of generating a predicted image signal using image signals of coded blocks adjacent to the block to be coded, and a coding step of coding block division information of the block to be coded, the block division step dividing the block to be coded in the recursive division into four blocks in the horizontal and vertical directions to generate four blocks.
The method includes a division step, and a bisection step of dividing a target block in the recursive division into two blocks horizontally or vertically to generate two blocks, and the bisection step prohibits dividing the target block for the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division if the previous recursive division was a bisection.

Yet another aspect of the present invention is an image decoding device, which performs decoding on a block basis into which an image is divided, and includes a decoding unit that decodes block division information of the blocks into which the image is divided, a block division unit that generates a current block to be decoded based on the decoded recursive block division information, and an intra prediction unit that generates a predicted image signal using an image signal of a decoded block adjacent to the current block to be decoded, the block division unit including a quartering unit that divides the current block in the recursive division into four blocks in the horizontal and vertical directions to generate four blocks, and a bisection unit that divides the current block in the recursive division into two blocks in the horizontal or vertical direction to generate two blocks, and the bisection unit prohibits, if the previous recursive division is a bisection, dividing the current block in the recursive division in the same direction as the direction in which the block was divided in the previous recursive division.

Yet another aspect of the present invention is an image decoding method, which performs decoding on a block-by-block basis into which an image is divided, and includes a decoding step of decoding block division information of the blocks into which the image is divided, a block division step of generating a current block to be decoded based on the decoded recursive block division information, and an intra prediction step of generating a predicted image signal using image signals of decoded blocks adjacent to the current block to be decoded, and the block division step includes a quartering step of dividing the current block in the recursive division into four horizontally and vertically to generate four blocks, and a bifurcating step of dividing the current block in the recursive division into two horizontally or vertically to generate two blocks.
a splitting step, the bisection step including, if the previous recursive splitting is a bisection,
It is prohibited to divide the target block of the current recursive division in the same direction as the block was divided in the previous recursive division.

Yet another aspect of the present invention is an image decoding program for decoding an image in units of blocks obtained by dividing the image, the program causing a computer to execute a decoding step of decoding block division information of the blocks obtained by dividing the image, a block division step of generating a current block to be decoded based on the recursively decoded block division information, and an intra prediction step of generating a predicted image signal using image signals of already-decoded blocks adjacent to the current block to be decoded, the block division step comprising:
The method includes a four-division step of dividing a target block in the recursive division into four blocks in the horizontal and vertical directions to generate four blocks, and a two-division step of dividing a target block in the recursive division into two blocks in the horizontal or vertical direction to generate two blocks, the two-division step including:
If the previous recursive division was a two-part division, it is prohibited to divide the target block of the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division.

Any combination of the above components, and any transformation of the present invention into a method, device, system, recording medium, computer program, etc., are also effective as aspects of the present invention.

According to the present invention, block division suitable for image encoding and decoding becomes possible, encoding efficiency is improved, and image encoding and decoding with a small amount of processing can be provided.

1 is a configuration diagram of an image encoding device according to a first embodiment. FIG. 1 is a configuration diagram of an image decoding device according to a first embodiment. 11 is a flowchart illustrating division into tree blocks and division within tree blocks. FIG. 13 is a diagram showing how an input image is divided into tree blocks. FIG. 1 is a diagram illustrating z-scanning. This is a diagram showing a tree block divided into four horizontally and vertically. This is a diagram showing a tree block divided horizontally into two. This is a diagram showing a tree block divided into two vertically. 13 is a flowchart for explaining the processing of each divided block when a tree block is divided into four in the horizontal and vertical directions. 13 is a flowchart illustrating processing of each divided block when a tree block is divided into two in the horizontal direction. FIG. 13 is a diagram showing how a divided block is re-divided when a tree block is divided into two horizontally. 13 is a flowchart illustrating processing of each divided block when a tree block is divided into two vertically. FIG. 13 is a diagram showing how a divided block is re-divided when a tree block is divided into two vertically. FIG. 4 illustrates an example of syntax related to block division according to the first embodiment. FIG. 1 is a diagram illustrating intra prediction. FIG. 1 is a diagram illustrating inter prediction. FIG. 13 illustrates an example of syntax related to block division according to the second embodiment. FIG. 13 is a diagram illustrating another example of the syntax related to block division according to the second embodiment. FIG. 13 illustrates an example of syntax related to block division according to the third embodiment. FIG. 13 is a diagram showing how the inside of a block that has been divided into two in the horizontal or vertical direction is further divided in the same direction. FIG. 13 is a diagram illustrating an example of syntax related to block division according to the fourth embodiment. FIG. 13 is a diagram showing the division of a tree block into four parts when the tree block is divided into two parts.

In the embodiment of the present invention, an image is divided into rectangular blocks, and the divided blocks are encoded and
The present invention provides an image coding technique for decoding.

(First embodiment)
An image coding device 100 and an image decoding device 200 according to a first embodiment of the present invention will be described. In the first embodiment, when block division is performed recursively, continuous division in the same direction is restricted.

Fig. 1 is a configuration diagram of an image coding device 100 according to a first embodiment. Fig. 1 shows only the flow of data related to image signals, and does not show the flow of data related to additional information, which is supplied by each component to an encoded bitstream generator 105 to generate corresponding encoded data, such as motion vectors and prediction modes.

The block division unit 101 divides an image into coding target blocks, which are processing units for coding, and supplies image signals in the coding target blocks to the residual signal generation unit 103. The block division unit 101 also supplies image signals of the coding target blocks to the prediction image generation unit 102 in order to evaluate the degree of matching of the prediction image.

The block division unit 101 recursively divides an image into rectangles of a predetermined size to generate a block to be coded. The block division unit 101 includes a quartering unit that divides the target block in the recursive division into four blocks in the horizontal and vertical directions to generate four blocks, and a bisection unit that divides the target block in the recursive division into two blocks in the horizontal or vertical direction to generate two blocks. The detailed operation of the block division unit 101 will be described later.

The predicted image generating unit 102 generates a predicted image based on a prediction mode from the decoded image signal supplied from the decoded image memory 108 by intra-picture prediction (intra prediction) or inter-picture prediction (inter prediction).
and generates a predicted image signal. The image signal of the block to be coded, supplied from the block division unit 101, is used for evaluation of intra prediction and inter prediction. In intra prediction, a predicted image signal is generated using the image signal of the block to be coded, supplied from the block division unit 101, and image signals of surrounding coded blocks adjacent to the block to be coded that exist in the same picture as the block to be coded, supplied from the decoded image memory 108. In inter prediction, the image signal of the block to be coded, supplied from the block division unit 101, is used as a reference picture using a coded picture stored in the decoded image memory 108 that is chronologically preceding or succeeding the picture (coded picture) that includes the block to be coded,
The prediction image generator 102 performs block matching between the coded picture and the reference picture, obtains a motion vector indicating the amount of motion, and performs motion compensation from the reference picture based on the amount of motion to generate a prediction image signal. The prediction image generator 102 supplies the prediction image signal thus generated to the residual signal generator 103.

The residual signal generating unit 103 performs subtraction between the image signal to be coded and the prediction signal generated by the prediction image generating unit 102 to generate a residual signal, and supplies the residual signal to the orthogonal transform and quantization unit 104 .

The orthogonal transform and quantization unit 104 orthogonally transforms and quantizes the residual signal supplied from the residual signal generation unit 103, and supplies the orthogonally transformed and quantized residual signal to the coded bit string generation unit 105 and the inverse quantization and inverse orthogonal transform unit 106.

The coded bit string generator 105 generates a coded bit string for the orthogonally transformed and quantized residual signal supplied from the orthogonal transform and quantization unit 104.
The unit 05 generates a corresponding encoded bit string for additional information such as a motion vector, a prediction mode, and block division information.

The inverse quantization/inverse orthogonal transform unit 106 inverse quantizes/inverse orthogonal transforms the orthogonally transformed/quantized residual signal supplied from the orthogonal transform/quantization unit 104, and supplies the inverse quantized/inverse orthogonal transformed residual signal to a decoded image signal superimposition unit 107.

The decoded image signal superimposition unit 107 superimposes the predicted image signal generated by the predicted image generation unit 102 and the residual signal that has been inverse quantized and inverse orthogonal transformed by the inverse quantization and inverse orthogonal transformation unit 106 to generate a decoded image, which is stored in the decoded image memory 108. Note that the decoded image may also be subjected to a filtering process for reducing block distortion and the like caused by encoding before being stored in the decoded image memory 108.

Fig. 2 is a configuration diagram of an image decoding device 200 according to embodiment 1. Fig. 2 shows only the flow of data related to image signals, and does not show the flow of data related to additional information other than image signals, such as motion vectors and prediction modes, which are supplied by a bitstream decoding unit 201 to each component and used for corresponding processing.

The bit string decoding unit 201 decodes the supplied coded bit string, and supplies the orthogonally transformed and quantized residual signal to the block division unit 202 .

The block division unit 202 determines the shape of the block to be decoded based on the decoded block division information, and supplies the orthogonally transformed and quantized residual signal of the determined block to be decoded to the inverse quantization and inverse orthogonal transformation unit 203.

The block division unit 202 recursively divides an image into rectangles of a predetermined size based on the decoded block division information to generate a block to be decoded. The block division unit 202 includes a quartering unit that divides the target block in the recursive division into four blocks in the horizontal and vertical directions to generate four blocks, and a bisection unit that divides the target block in the recursive division into two blocks in the horizontal or vertical direction to generate two blocks. The detailed operation of the block division unit 202 will be described later.

The inverse quantization and inverse orthogonal transformation unit 203 performs inverse orthogonal transformation and inverse quantization on the orthogonally transformed and quantized residual signal supplied thereto, thereby obtaining an inverse orthogonally transformed and inverse quantized residual signal.

The predicted image generating unit 204 generates a predicted image signal from the decoded image signal supplied from the decoded image memory 206 , and supplies the generated predicted image signal to the decoded image signal superimposing unit 205 .

The decoded image signal superimposition unit 205 generates a decoded image signal by superimposing the predicted image signal generated by the predicted image generation unit 204 and the residual signal that has been inverse orthogonally transformed and inverse quantized by the inverse quantization and inverse orthogonal transformation unit 203, and outputs the decoded image signal while storing it in the decoded image memory 206.
The decoded image may be subjected to a filtering process to reduce block distortion and the like caused by encoding, and then stored in the decoded image memory 206 .

A detailed description will now be given of the operation of the block division unit 101 of the image encoding device 100. Fig. 3 is a flowchart for explaining division into tree blocks and division inside a tree block.

First, an input image is divided into tree blocks of a predetermined size (S1000). For example, a tree block is 128 pixels x 128 pixels.
The size and shape of the tree block are not limited to 128 pixels by 128 pixels, and any size and shape may be used as long as it is rectangular. The size and shape of the tree block may be fixed between the encoding device and the decoding device, or may be determined by the encoding device and recorded in the encoded bit stream, and the decoding device may use the recorded block size. Figure 4 shows how an input image is divided into tree blocks. The tree blocks are encoded and decoded in raster scan order, i.e., from left to right and from top to bottom.

The inside of the tree block is further divided into rectangular blocks. The inside of the tree block is coded and decoded in the order of z-scanning. Figure 5 shows the order of z-scanning. In z-scanning, coding and decoding are performed in the order of top left, top right, bottom left, and bottom right. The inside of the tree block can be divided into four or two, and the four divisions are made both horizontally and vertically. 2
The division is performed horizontally or vertically. FIG. 6 shows a tree block divided into four blocks horizontally and vertically.
Fig. 7 is a diagram of a tree block divided into two in the horizontal direction. Fig. 8 is a diagram of a tree block divided into two in the vertical direction.

Referring again to Fig. 3, it is determined whether the inside of the tree block is divided into four horizontally and vertically (S1001).

If it is determined that the inside of the tree block should be divided into four (S1001: Yes), the inside of the tree block is divided into four (S1002), and each of the blocks divided into four horizontally and vertically is processed (S1003). The re-division process of the block divided into four will be described later (FIG. 9).

If it is determined that the inside of the tree block should not be divided into four (S1001: No), it is determined whether or not to divide the inside of the tree block into two (S1004).

If it is determined that the inside of the tree block is to be divided into two (S1004: Yes), it is determined whether the direction of division into two is the horizontal direction (S1005).

If it is determined that the direction of division into two is horizontal (S1005: Yes), the inside of the tree block is divided into two horizontally (S1006), and each process of the blocks divided into two horizontally is performed (S1007). The re-division process of the blocks divided into two horizontally will be described later ( FIG. 10 ).

If it is determined that the direction of division is vertical rather than horizontal (S1005: No), the inside of the tree block is divided vertically into two (S1008), and each process of the blocks divided vertically into two is performed (S1009). The re-division process of the block divided horizontally into two will be described later ( FIG. 11 ).

If it is determined that the inside of the tree block is not to be divided into two (S1004: No), the block division process ends without dividing the inside of the tree block into blocks (S1010).

Next, the processing of each of the divided blocks when the tree block is divided into four in the horizontal and vertical directions will be described with reference to the flowchart of FIG.

It is judged whether the inside of the block is divided again into four parts horizontally and vertically (S1101).
.

If it is determined that the inside of the block should be divided into four again (S1101: Yes), the inside of the block is divided into four again (S1102), and each of the blocks divided into four in the horizontal and vertical directions is processed (S1103).

If it is determined that the inside of the block is not to be divided into four again (S1101: No), it is determined whether or not to divide the inside of the block into two (S1104).

If it is determined that the inside of the block is to be divided into two (S1104: Yes), it is determined whether the direction of division into two is the horizontal direction (S1105).

If it is determined that the direction of division into two is the horizontal direction (S1105: Yes), the inside of the block is divided into two in the horizontal direction (S1106), and each process of the blocks divided into two in the horizontal direction is performed (S1107).
107).

If it is determined that the direction of division is vertical rather than horizontal (S1105: No), the block is divided vertically into two (S1108), and each process of the blocks divided vertically into two (S1109) is performed.

If it is determined that the inside of the block is not to be divided into two (S1104: No), the block division process ends without dividing the inside of the block into blocks (S1110).

9 is executed for each of the four divided blocks. Encoding and decoding are also performed within each of the four divided blocks in the z-scan order.

Next, the processing of each divided block when a tree block is divided into two in the horizontal direction will be described with reference to the flowchart of FIG.

When a tree block is divided into two blocks in the horizontal direction, first, for each of the divided blocks, it is determined whether or not the inside of the block should be divided into four blocks in the horizontal and vertical directions (S1201).

If it is determined that the inside of the block is to be divided into four (S1201: Yes),
The image is divided into four blocks (S1202), and each block is processed (S12
03).

If it is determined that the inside of the block is not to be divided into four (S1201: No), it is determined whether or not to divide the inside of the block into two again (S1204).

If it is determined that the block should be divided into two again (S1204: Yes), the inside of the block is divided vertically (S1205), and each process of the blocks divided vertically into two is carried out (S1206).

If it is determined that the block is not to be divided into two again (S1204: No), the block division process ends without dividing the inside of the block again (S1207).

Fig. 11 shows how a divided block is re-divided when a tree block is divided into two horizontally. Here, when a parent tree block is divided into two horizontally, the re-division of the divided block is permitted only in the vertical direction, and the block is automatically divided into two vertically. Also, when a parent tree block is divided into two, it is possible to completely prohibit the child block from being divided into four. This makes it possible to prohibit a block from being divided in the same direction as the parent block, thereby preventing block division into a horizontally elongated rectangle, and facilitating the coding and decoding process.

10 is executed for each of the blocks divided into two in the horizontal direction. The inside of each divided block is also coded and decoded in the order of top to bottom.

Next, the processing of each divided block when a tree block is divided into two vertically will be described with reference to the flowchart of FIG.

When a tree block is divided into two blocks vertically, each of the divided blocks first determines whether or not to divide the block into four blocks horizontally and vertically (S1301).

If it is determined that the inside of the block is to be divided into four (S1301: Yes),
The image is divided into four blocks (S1302), and each block is processed (S13
03).

If it is determined that the inside of the block is not to be divided into four (S1301: No), it is determined whether or not to divide the inside of the block into two again (S1304).

If it is determined that the block should be divided into two again (S1304: Yes), the block is divided horizontally (S1305), and each process is performed on the blocks divided into two horizontally (S1306).

If it is determined that the block is not to be divided into two again (S1304: No), the block division process ends without dividing the inside of the block again (S1307).

Fig. 13 shows how a divided block is re-divided when a tree block is divided into two vertically. Here, when a parent tree block is divided into two vertically, the divided block is automatically divided into two horizontally, allowing only horizontal division. Also, when a parent tree block is divided into two, it is possible to completely prohibit a child block from being divided into four. This makes it possible to prohibit a block from being divided in the same direction as the parent block, thereby preventing block division into a vertically elongated rectangle, and facilitating coding and decoding processes.

12 is executed for each of the blocks divided into two vertically. The inside of each divided block is also coded and decoded in the order of left, right.

Although the re-division of the divided blocks when the tree block is divided has been described, the parent block does not have to be a tree block. For example,
28) is divided into four, and when each of the four divided blocks (64x64) is further divided into four or into two, the above process is also applied to the division of the further divided blocks.

Next, the operation of the block division unit 202 of the image decoding device 200 will be described. Blocks are divided in the same manner as the block division unit 101 of the image encoding device 100. However, while the block division unit 101 of the image encoding device 100 selects a block division pattern and outputs the selected block division information, the block division unit 202 of the image decoding device:
The difference is that blocks are divided using block division information decoded from the encoded bitstream, and that when decoding the block division information from the encoded bitstream, in a situation where re-division in the same direction is prohibited, information that is not an option is not transmitted in the bitstream using a syntax structure.

An example of the syntax (syntax rules for encoded bit streams) related to block division in the first embodiment is shown in Fig. 14. When dividing the inside of a tree block, a flag (4_division_flag) indicating whether or not to divide into four is first transmitted and received. When dividing into four (4_division_flag is 1),
In the case of the tree block, the tree block is divided into four parts and the process is terminated. After that, the tree block is divided into four parts again using the syntax shown in FIG. 14.
If lag is 0, a flag (2_division_flag) indicating whether to divide into two is sent and received. If dividing into two (2_division_flag is 1), a flag (2_division_d) indicating the direction of division is also sent and received.
2_division_direction is 1, which indicates vertical division, and
If _division_direction is 0, it indicates division in the horizontal direction. After that, the block divided into two is re-divided inside the block using the syntax shown in Fig. 14. If the block is not to be divided into two (2_division_flag is 0), the process ends without dividing the tree block.

Here, a process for further dividing the inside of a block divided into four or two will be described.
The process of re-dividing the inside of a block also uses the syntax shown in Fig. 14, but is different from the case of dividing a tree block in that there is a restriction on the division direction when dividing into two. In other words, when a tree block is divided into two, when the inside of the divided block is to be re-divided, it is prohibited to divide the tree block in the same direction as the division direction when the tree block was divided into two.
This makes it possible to prevent the divided blocks from becoming elongated rectangles, and to prevent an increase in memory bandwidth required for intra prediction and inter prediction. Prevention of an increase in memory bandwidth will be described in detail later.

Alternatively, the number of times the image is divided into two in the same direction may be counted, and if the number of times exceeds a predetermined number, further division in the same direction may be restricted. For example, division into two in the same direction may be permitted up to two times, but division into two in the same direction may be prohibited from the third time onward.

In Fig. 14, the syntax is such that division into four parts is given priority, and information on whether to divide into four parts is transmitted and received before information on whether to divide into two parts. On the other hand, when division into two parts is given priority, it is also possible to use a syntax in which information on whether to divide into two parts is transmitted and received before information on whether to divide into four parts. This is because the amount of code transmitted as a bit stream is reduced by transmitting and receiving an event that is more likely to occur stochastically first. In other words, if division into four parts and division into two parts are selected in advance,
It is also possible to use a syntax that estimates which of the two divisions is more likely to occur and transmits and receives division information that is more likely to occur first. For example, by transmitting and receiving information in the header information of an image indicating whether to prioritize division into four or two, the encoding device can adaptively determine the number of divisions with the highest encoding efficiency, and the decoding device can divide the inside of a tree block using a syntax based on the selected number of divisions with the highest priority.

In the image encoding device 100 and the image decoding device 200, intra prediction and inter prediction are performed using the divided blocks. Both intra prediction and inter prediction involve copying pixels from memory.

An example of intra prediction is shown in FIG. 15(a) to FIG. 15(d).
b) indicates the prediction direction and mode number of intra prediction. As shown in FIG. 15(c) and FIG. 15(d), intra prediction generates a predicted image of a block to be coded/decoded by copying pixels from pixels that have already been coded/decoded and are adjacent to the block to be coded/decoded. In intra prediction, the process from the generation of a predicted image to the generation of coded/decoded pixels is repeated for each block, so the processing order is sequential for each block, and the more the inside of the block is divided into smaller parts, the greater the overall processing load. Also, the more elongated the shape of the block is, the greater the processing of copying pixels from memory becomes. Also, since orthogonal transformation of residual signals is performed for coding/decoding, the more types of rectangular sizes there are, the more types of orthogonal transformation are required, which results in an increase in the circuit scale. Therefore, when the inside of a block is divided into two,
By restricting the division into two in the same direction as the division method of the parent block, it is possible to prevent an increase in memory bandwidth required for intra prediction.

FIG. 16 shows an example of inter prediction. In inter prediction, a predicted image of a block to be coded/decoded is generated by copying pixels from pixels included in a coded/decoded image in block units. In inter prediction, when copying pixels from a reference image in block units, the device configuration often requires acquisition of memory management units containing necessary pixels. Therefore, the smaller the block is divided, and the more elongated the shape of the block is, the heavier the overall processing load becomes. In addition, when performing decimal precision motion compensation using an interpolation filter on a reference image, it is necessary to copy pixels by adding several pixels to the pixels included in the block, and the smaller the size of the block, the larger the relative ratio of the several pixels to be added becomes, and the heavier the overall processing load becomes. Therefore, when dividing the inside of a block into two, by restricting the division into two in the same direction as the division direction of the parent block, it is possible to prevent an increase in the memory bandwidth required for inter prediction.

Second Embodiment
An image coding device and an image decoding device according to a second embodiment of the present invention will be described. The second embodiment differs from the first embodiment in that it restricts further division of a block when the block is equal to or smaller than a predetermined size, but the rest of the configuration is the same as the first embodiment. This prevents the overall processing load from increasing as the block is divided into smaller parts.

17 and 18 show the syntax for block division according to the second embodiment.
14 in the embodiment is that block division is possible only when the size of the block is initially larger than a predetermined size. In the case of Fig. 17, when the number of pixels in a block is greater than 64, the block can be divided into four or two.

Furthermore, when taking into consideration the difference in the number of pixels in a block divided by 4 division and 2 division, as shown in Fig. 18, 4 division is permitted when the number of pixels in the block is greater than 64, and 2 division is permitted when the number of pixels in the block is greater than 32. This makes it possible to precisely control the limit on the number of pixels in the divided blocks.

Third Embodiment
An image coding device and an image decoding device according to a third embodiment of the present invention will be described. The third embodiment is different from the first embodiment in that a block divided in the vertical direction is restricted from being further divided in the vertical direction, but other configurations are the same as those of the first embodiment.

Typically, pixel information of an image is stored in a one-dimensional memory in raster scan order. That is, in the one-dimensional memory, pixels in the horizontal direction are stored relatively close to each other, and pixels in the vertical direction are stored relatively far away. Therefore, it is easy to access pixels in the horizontal direction, but it is not easy to access pixels in the vertical direction. For example, in the case of a block of 16 horizontal pixels by 8 vertical pixels and a block of 8 horizontal pixels by 16 vertical pixels, the number of pixels is the same, but the block of 8 horizontal pixels by 16 vertical pixels has a larger memory range in which pixels are stored than the block of 16 horizontal pixels by 8 vertical pixels. Therefore, more memory bandwidth is required for pixel transfer when using motion compensation.

Figure 19 shows the syntax for block division in the third embodiment. The difference from the syntax in Figure 14 in the first embodiment is that only when a parent block is divided into two vertically, further division into two vertically is prohibited. That is, as shown in Figure 20,
When a parent block is vertically divided into two, and the inside of the parent block is further divided into two, the horizontal division is automatically selected without the need to select between the horizontal and vertical divisions.

(Fourth embodiment)
An image coding device and an image decoding device according to a fourth embodiment of the present invention will be described. The fourth embodiment is different from the first embodiment in that, when a block is divided into two and then an inner block is divided into four, the inner block divided into four is prohibited from being further divided. The other configurations are the same as those of the first embodiment.

21 shows the syntax for block division in the fourth embodiment. As shown in FIG. 22, when a parent block is divided into two and the internal blocks are divided into four, 2_di
vision_after_4_division_flag becomes 1, disallowing all subsequent divisions.

This is because, when dividing into two and then dividing into four, it is confirmed that dividing into two was not selected when dividing into two, so there is a low possibility that further block division will be required after dividing into two and then dividing into four. In such a case, dividing into four should be selected from the beginning.
When a block is divided into two and then divided into four, the shape of the block is already rectangular, so if it is attempted to prohibit further division into two after the four division, the processing becomes complicated. If subsequent block division is uniquely prohibited for blocks divided into four after the two division, the processing for determining whether block division is possible does not become complicated. By uniquely prohibiting subsequent block division for blocks divided into four after the two division, it becomes unnecessary to transmit and receive the selection of not dividing the block in the bit stream, and the amount of code to be transmitted can be reduced.

Of course, it is also possible to combine a plurality of block division restriction techniques according to the first to fourth embodiments.

The image coding bit stream output by the image coding device according to the embodiment described above is
It has a specific data format so that it can be decoded according to the encoding method used in the embodiment, and an image decoding device corresponding to the image encoding device can decode the encoded bit stream of this specific data format.

When a wired or wireless network is used to exchange coded bitstreams between an image coding device and an image decoding device, the coded bitstream may be converted into a data format suitable for the transmission mode of the communication channel and then transmitted. In this case, a transmitting device is provided that converts the coded bitstream output by the image coding device into coded data in a data format suitable for the transmission mode of the communication channel and transmits the coded bitstream to the network, and a receiving device is provided that receives the coded data from the network, restores it to a coded bitstream, and supplies it to the image decoding device.

The transmitting device includes a memory that buffers the coded bit stream output by the image coding device, a packet processing unit that packetizes the coded bit stream, and a transmitting unit that transmits the packetized coded data via a network. The receiving device includes a receiving unit that receives the packetized coded data via the network, a memory that buffers the received coded data, and a packet processing unit that packetizes the coded data to generate a coded bit stream and provides it to the image decoding device.

In addition, by adding a display unit for displaying the image decoded by the image decoding device,
It is also possible to use it as a display device. In this case, the display unit reads out the decoded image signal generated by the decoded image signal superimposing unit 205 and stored in the decoded image memory 206, and displays it on a screen.

In addition, it is possible to configure the image encoding device by adding an imaging unit and inputting a captured image to the image encoding device. In this case, the imaging unit inputs a captured image signal to the block division unit 101.

The above-mentioned encoding and decoding processes can be realized not only as a transmission, storage, and receiving device using hardware, but also as firmware stored in a ROM (read only memory) or flash memory, or as software for a computer, etc. The firmware and software programs can be provided by recording them on a recording medium readable by a computer, etc., or can be provided from a server via a wired or wireless network, or can be provided as data broadcasting of terrestrial or satellite digital broadcasting.

The present invention has been described above based on the embodiments. The embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each treatment process, and that such modifications are also within the scope of the present invention.

100 Image encoding device, 101 Block division unit, 102 Prediction image generation unit,
103 Residual signal generation unit, 104 Orthogonal transformation and quantization unit, 105 Encoded bit string generation unit, 106 Inverse quantization and inverse orthogonal transformation unit, 107 Decoded image signal superimposition unit, 108
Decoded image memory, 200 Image decoding device, 201 Bit stream decoding unit, 202 Block division unit, 203 Inverse quantization and inverse orthogonal transformation unit, 204 Prediction image generation unit, 20
5 Decoded image signal superimposing unit, 206 Decoded image memory.

Claims (2)

A method for storing a bit stream generated by a moving image encoding method on a recording medium, the method comprising the steps of:
a block division step of recursively dividing an image into rectangles of a predetermined size to generate blocks to be encoded;
An encoding step of encoding block division information of a block to be encoded,
The block division step includes:
a 4-division step for dividing the target block in the recursive division into 4 blocks horizontally and vertically to generate 4 blocks;
A bisection step of bisecting the target block in the recursive division in a horizontal or vertical direction to generate two blocks,
The four blocks are coded in the following order: top left, top right, bottom left, bottom right.
In the bisection step, when the previous recursive division was bisection and the block of the encoding target block is equal to or smaller than a predetermined size, the bisection step prohibits dividing the target block of the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division.
A storage method comprising: A transmission method for transmitting a bitstream generated by a video encoding method, the video encoding method comprising:
a block division step of recursively dividing an image into rectangles of a predetermined size to generate blocks to be encoded;
An encoding step of encoding block division information of a block to be encoded,
The block division step includes:
a 4-division step for dividing the target block in the recursive division into 4 blocks horizontally and vertically to generate 4 blocks;
A bisection step of bisecting the target block in the recursive division in a horizontal or vertical direction to generate two blocks,
The four blocks are coded in the following order: top left, top right, bottom left, bottom right.
In the bisection step, when the previous recursive division was bisection and the block of the encoding target block is equal to or smaller than a predetermined size, the bisection step prohibits dividing the target block of the current recursive division in the same direction as the direction in which the block was divided in the previous recursive division.
A transmission method comprising:

Images