JP6315137B2 - Moving picture decoding apparatus, moving picture decoding method, and moving picture decoding program - Google Patents

Description

The present invention relates to a moving picture decoding technique, and more particularly to a moving picture decoding technique using predictive coding of quantization parameters.

MPEG-2 Part 2 (hereinafter referred to as MPEG-2) and MPEG-4 Part 10
/ H. In digital video coding such as H.264 (hereinafter referred to as AVC), an image is divided into blocks of a predetermined size and coded to indicate the roughness of quantization for a prediction error signal (or simply an image signal). Transmit quantization parameters. By variably controlling the quantization parameter in units of predetermined blocks on the encoding side, it is possible to control the code amount and improve the subjective image quality.

As a control of the quantization parameter for improving the subjective image quality, Adaptive Quan
Tization (adaptive quantization) is often used. In adaptive quantization, it is changed according to the activity of each macroblock so that it is quantized more finely in the flat part that is visually noticeable, and coarser in the complicated part of the pattern that is relatively inconspicuous. . That is, in a macroblock with a high activity that tends to have a large allocated bit amount when encoded, the quantization parameter is changed so that a large quantization scale is set. Subjective image quality is improved while controlling the number of bits to be as small as possible in the data.

In MPEG-2, it is determined whether the quantization parameter of the previous block and the quantization parameter of the block to be encoded are the same in the encoding / decoding order, and if not, the quantization parameter is transmitted. To do. In AVC, the quantization parameter of the previous block in the encoding / decoding order is used as a prediction value, and the quantization parameter of the encoding target block is differentially encoded. This is based on the fact that since the code amount control is generally performed in the coding order, the quantization parameter of the previous block in the coding order is closest to the quantization parameter of the coding block. It aims to suppress the amount of information of the conversion parameter.

JP 2011-91772 A

In the conventional quantization parameter control, the difference between the quantization parameter of the encoding target block is calculated using the quantization parameter of the encoded left block as a prediction quantization parameter, and the calculated difference quantization parameter is calculated. The amount of code of the quantization parameter was reduced by encoding. However, depending on the content in the screen, for example, as shown in FIG. 8, when the characteristics of the image in the encoding target block and the image in the encoded left block are different, it is calculated by adaptive quantization. Since the difference between the quantization parameters becomes large, even if the quantization parameter prediction with the left block is performed uniquely, the difference quantization parameter becomes large,
There was a problem that the amount of codes increased.

In addition, since the quantization parameter calculated by the code amount control is performed in the raster scan order from the upper left to the lower right of the normal screen, the processing order is separated between slices when the block size to be encoded is reduced. Therefore, when the quantization parameter of the encoded block that is close to the encoding target block is used for prediction, it is close but the processing order in the code amount control is far away, so the code amount control The quantization parameter calculated in step 1 is not necessarily the same or close to the encoding target block and the encoded block adjacent to the upper side, and it cannot be said that the code amount of the differential quantization parameter can be reduced. There was a problem.

The present invention has been made in view of such circumstances, and an object of the present invention is to provide a technique for improving the coding efficiency by reducing the code amount of the quantization parameter.

In order to solve the above problems, a moving picture decoding apparatus according to an aspect of the present invention further divides a first block obtained by dividing each picture of a moving picture by a predetermined size into one or a plurality of second blocks. Then, a moving picture decoding apparatus for decoding a bit stream in which the moving picture is encoded, the block size information for decoding the bit stream and deriving a prediction quantization parameter, and the second block Using the decoding unit for extracting the differential quantization parameter, the quantization parameter of the third block adjacent to the left of the second block and the quantization block of the fourth block adjacent to the second block; A predictive quantization parameter deriving unit for deriving a predictive quantization parameter of the second block; and addition of the differential quantization parameter of the second block and the predictive quantization parameter A predictive quantization parameter deriving unit at a position where the third block does not exceed the boundary of the first block. In some cases, the quantization parameter of the third block is set as the first quantization parameter, and when there is a position beyond the boundary of the first block, it is decoded immediately before the second block in decoding order. If the quantization parameter of the fifth block is the first quantization parameter, and the fourth block is at a position not exceeding the boundary of the first block, the quantization parameter of the fourth block The second
The quantization parameter of the fifth block is set as the second quantization parameter for each predetermined size based on the size information. The first quantization parameter and the second quantization parameter are used to derive the predicted quantization parameter of the second block, and the decoding unit has a shorter code length as the absolute value is smaller. The differential quantization parameter of the second block is decoded.

Yet another aspect of the present invention is a video decoding method. In this method, a first block obtained by dividing each picture of a moving image by a predetermined size is further divided into one or a plurality of second blocks, and a moving picture for decoding a bitstream in which the moving image is encoded is decoded. An image decoding method comprising:
A decoding step of extracting block size information for decoding the bitstream and deriving a predictive quantization parameter and a differential quantization parameter of the second block; and a third step adjacent to the left of the second block And a fourth adjacent to the second block and the second block
A prediction quantization parameter deriving step for deriving a prediction quantization parameter of the second block using a quantization parameter of the block of the second block, and addition of the difference quantization parameter of the second block and the prediction quantization parameter A quantization parameter generation step of generating a quantization parameter of the second block by the step, wherein the predicted quantization parameter derivation step includes a position where the third block does not exceed the boundary of the first block. If the quantization parameter of the third block is the first quantization parameter, and the position exceeds the boundary of the first block, decoding is performed immediately before the second block in decoding order. The quantization parameter of the fifth block thus set is set as the first quantization parameter, and the fourth block is the boundary of the first block. The fourth block quantization parameter is set as the second quantization parameter, and when the position is beyond the boundary of the first block, the fourth block quantization parameter A quantization parameter is set as a second quantization parameter, and the first quantization parameter and the second quantization parameter are set for each predetermined size based on the size information.
Is derived as a predictive quantization parameter of the second block, and the decoding step decodes the differential quantization parameter of the second block so that the code length becomes shorter as the absolute value becomes smaller .

Yet another aspect of the present invention is a moving image decoding program. This program is a moving image that decodes a bit stream in which the moving picture is encoded by dividing the first block obtained by dividing each picture of the moving picture into a predetermined size into one or a plurality of second blocks. A decoding step for extracting block size information for decoding the bitstream and deriving a predictive quantization parameter and a differential quantization parameter of the second block; Prediction quantization parameter derivation for deriving the prediction quantization parameter of the second block using the quantization parameter of the third block adjacent to the left of the second block and the quantization block of the fourth block adjacent to the second block And adding the difference quantization parameter of the second block and the prediction quantization parameter of the second block A quantization parameter generation step of generating a child parameter, and when the predicted quantization parameter derivation step is at a position where the third block does not cross the boundary of the first block, If the quantization parameter of the third block is the first quantization parameter and the position is beyond the boundary of the first block, the fifth block decoded immediately before the second block in decoding order If the quantization parameter of the block is the first quantization parameter and the fourth block is in a position not exceeding the boundary of the first block, the quantization parameter of the fourth block is set to the second quantization parameter. If the quantization parameter is located beyond the boundary of the first block, the quantization parameter of the fifth block is set to the second quantization parameter. Deriving as a predicted quantization parameter of the second block using the first quantization parameter and the second quantization parameter for each predetermined size based on the size information, and the decoding step Decodes the differential quantization parameter of the second block so that the smaller the absolute value, the shorter the code length.

It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

According to the present invention, it is possible to improve the coding efficiency by reducing the code amount of the quantization parameter.

It is a block diagram which shows the structure of the moving image encoder which comprised the derivation | leading-out method of the prediction quantization parameter which concerns on embodiment. It is a block diagram which shows the structure of the moving image decoding apparatus provided with the derivation method of the prediction quantization parameter which concerns on embodiment. It is a figure explaining the code amount control in the screen of MPEG-2 TM5. H. It is a figure which shows the H.264 quantization parameter prediction method. It is a figure which shows an example of the encoding process order at the time of using hierarchy tree encoding. It is a figure which shows prediction of the quantization parameter of the encoding block of the upper left inside the tree block divided | segmented by hierarchy tree encoding. It is a figure which shows an example of the encoding process order inside the tree block divided | segmented by hierarchy tree encoding. It is a figure showing an example with which a design is contained in the block on the left and the upper left with respect to the surrounding block which has already been encoded close to the encoding block of encoding object. FIG. 4 is a diagram for explaining the positions of adjacent coding blocks in the upper and lower sides in code amount control within a screen of MPEG-2 TM5. It is a block diagram which shows the detailed structure of the prediction quantization parameter derivation | leading-out part of this Embodiment. It is a figure explaining the value and prediction direction of the intra prediction mode prescribed | regulated by this Embodiment. It is a figure showing an example of a Golomb encoding table with a signed exponent sign of a difference quantization parameter. It is a figure which shows the relationship between an encoding object tree block and an encoded tree block. It is a figure which shows the relationship between the encoding block inside the tree block divided | segmented by hierarchy tree encoding, and the encoded block. In a 1st Example, it is a figure showing the reference place of the prediction quantization parameter of an encoding block. It is a figure showing an example which shows as a reference destination the quantization parameter of the adjacent block of an already encoded near as a prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the 1st prediction quantization parameter derivation part of the 1st example. It is a flowchart for demonstrating another operation | movement of the 1st prediction quantization parameter derivation part of the 1st example. In a 2nd Example, it is a figure showing the reference destination of the prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the 1st prediction quantization parameter derivation part of the 2nd example. In a 3rd Example, it is a figure showing the reference place of the prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the 1st prediction quantization parameter derivation part of the 3rd example. In a 4th Example, it is a figure showing the reference destination of the prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the 1st prediction quantization parameter derivation part of the 4th example. In a 5th Example, it is a figure showing the reference place of the prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the prediction quantization parameter derivation | leading-out part of a 5th Example. It is a figure explaining the value and prediction direction of the intra prediction mode prescribed | regulated by this Embodiment. It is a flowchart for demonstrating operation | movement of the 2nd prediction quantization parameter derivation | leading-out part 304 of a 6th Example. It is a figure explaining the table which converts intra prediction mode into an intra prediction direction. It is a figure explaining the setting of the quantization parameter of the adjacent block referred in the prediction direction of intra prediction mode in a 6th Example. It is a flowchart for demonstrating the detailed operation | movement of the determination process by an intra prediction direction in the 2nd prediction quantization parameter derivation part of the 6th example. It is a figure explaining the setting of a weighting coefficient by the prediction direction of intra prediction mode in a 6th Example. It is a flowchart for demonstrating the detailed operation | movement of the determination process by an intra prediction direction in the 2nd prediction quantization parameter derivation part of the 6th example. It is a figure explaining the setting of another weighting coefficient by the prediction direction of intra prediction mode in a 6th Example. FIG. 35 is a flowchart for explaining detailed operation when the intra prediction direction falls within a predetermined range shown in FIG. 34 in the second predictive quantization parameter deriving unit of the sixth embodiment. In a 7th Example, it is a figure showing the reference place of the prediction quantization parameter of an encoding block. It is a flowchart for demonstrating operation | movement of the 2nd prediction quantization parameter derivation part of the 7th example. It is a flowchart for demonstrating the detailed operation | movement of the determination process by an intra prediction direction in the 2nd prediction quantization parameter derivation part of the 7th example. It is a flowchart for demonstrating operation | movement of the 2nd prediction quantization parameter derivation part of the 8th example. It is a flowchart for demonstrating the detailed operation | movement of the determination process by an intra prediction direction in the 2nd prediction quantization parameter derivation part of the 8th example. It is a flowchart for demonstrating detailed operation | movement when the intra prediction direction is 0 in the 2nd prediction quantization parameter derivation part of the 8th example. It is a flowchart for demonstrating a detailed operation | movement when the intra prediction direction is less than 18 in the 2nd prediction quantization parameter derivation part of the 8th example. It is a flowchart for demonstrating detailed operation | movement when the intra prediction direction is 18 or more in the 2nd prediction quantization parameter derivation part of the 8th example. It is a flowchart for demonstrating the detailed operation | movement which switches a weighted coefficient by the determination process in case the intra prediction direction is 0 in the 2nd prediction quantization parameter derivation part of an 8th Example. FIG. 35 is a flowchart for explaining a detailed operation of switching weighted coefficients in the determination process based on the intra prediction direction shown in FIG. 34 in the second predictive quantization parameter deriving unit of the eighth embodiment. It is a block diagram which shows another detailed structure of a prediction quantization parameter derivation | leading-out part. It is a flowchart for demonstrating operation | movement of the 2nd prediction quantization parameter derivation part of the 9th example. It is a flowchart for demonstrating the detailed operation | movement of the determination process by an intra prediction direction in the 2nd prediction quantization parameter derivation part of the 9th example. It is a flowchart for demonstrating detailed operation | movement when the intra prediction direction is 0 in the 2nd prediction quantization parameter derivation part of the 9th example. It is a flowchart for demonstrating detailed operation | movement when the intra prediction direction is less than 18 in the 2nd prediction quantization parameter derivation part of the 9th example. It is a flowchart for demonstrating detailed operation | movement when the intra prediction direction is 18 or more in the 2nd prediction quantization parameter derivation part of the 9th example. It is a flowchart for demonstrating the detailed operation | movement of a recalculation determination part. It is a figure explaining an example of a quantization group. It is a figure explaining an example of prediction of the quantization parameter of a quantization group unit.

The embodiment of the present invention divides a picture into rectangular blocks of a predetermined size, further divides the block into one or a plurality of encoded blocks, and performs video coding that performs quantization / encoding in units of encoded blocks In order to reduce the coding amount of the quantization parameter of the block to be processed, the optimal prediction quantization parameter is derived from the coding information of the surrounding coded blocks and the difference from the prediction quantization parameter is calculated. Thus, a code amount control technique for encoding is provided.

A preferred moving image encoding apparatus 100 and moving image decoding apparatus 200 that implement the present invention will be described. FIG. 1 is a block diagram showing the configuration of a moving picture coding apparatus 100 that implements the present invention, in which an image memory 101, a residual signal generation unit 102, an orthogonal transform / quantization unit 103, and a second encoded bit string generation unit. 104, an inverse quantization / inverse orthogonal transform unit 105, a decoded image signal superimposing unit 106, a decoded image memory 107, a predicted image generation unit 108, an activity calculation unit 109, a quantization parameter calculation unit 110, a differential quantization parameter generation unit 111, A first encoded bit string generation unit 112, an encoded information storage memory 113, a predicted quantization parameter derivation unit 114, and an encoded bit string multiplexing unit 115 are configured. The thick solid arrows connecting the blocks represent picture image signals, and the thin solid arrows represent the flow of parameter signals for controlling encoding.

The image memory 101 temporarily stores image signals to be encoded supplied in the order of shooting / display time. The image memory 101 supplies the stored image signal to be encoded to the residual signal generation unit 102, the predicted image generation unit 108, and the activity calculation unit 109 in units of predetermined pixel blocks. At this time, the images stored in the order of shooting / display time are rearranged in the encoding order and output from the image memory 101 in units of pixel blocks.

The residual signal generation unit 102 subtracts the image signal to be encoded from the prediction signal generated by the prediction image generation unit 108 to generate a residual signal, and supplies the residual signal to the orthogonal transform / quantization unit 103.

The orthogonal transform / quantization unit 103 performs orthogonal transform and quantization on the residual signal, generates an orthogonal transform / quantized residual signal, and performs inverse quantization with the second encoded bit string generation unit 104. Supply to the inverse orthogonal transform unit 105.

The second encoded bit sequence generation unit 104 generates a second encoded bit sequence by entropy encoding the orthogonally transformed and quantized residual signal according to a prescribed syntax rule, and the encoded bit sequence multiplexing unit 115 Supply.

The inverse quantization / inverse orthogonal transform unit 105 calculates a residual signal by performing inverse quantization and inverse orthogonal transform on the orthogonal transform / quantized residual signal supplied from the orthogonal transform / quantization unit 103 to perform decoding. This is supplied to the image signal superimposing unit 106.

The decoded image signal superimposing unit 106 superimposes the predicted image signal generated by the predicted image generating unit 108 and the residual signal subjected to inverse quantization and inverse orthogonal transform by the inverse quantization / inverse orthogonal transform unit 105 to generate a decoded image. It is generated and stored in the decoded image memory 107. The decoded image may be subjected to a filtering process for reducing distortion such as block distortion due to encoding, and may be stored in the decoded image memory 107. In that case, a post filter such as a deblocking filter may be used as necessary. Predicted encoded information such as a flag for identifying the information is stored in the encoded information storage memory 113.

The predicted image generation unit 108 uses an image signal supplied from the image memory 101 and a decoded image signal supplied from the decoded image memory 107, based on a prediction mode, for intra-frame prediction (intra prediction) or inter-frame prediction (inter prediction). To generate a predicted image signal. Intra prediction is a coding target block obtained by dividing an image signal supplied from the image memory 101 by a predetermined block unit and a coding target block supplied from the decoded image memory 107 in the same frame. A prediction image signal is generated using pixel signals of surrounding encoded blocks adjacent to the block to be converted. In the inter prediction, the decoded image memory 107 is separated by several frames before or behind in the time series of the frames (encoded frames) of the block to be encoded obtained by dividing the image signal supplied from the image memory 101 in a predetermined block unit. The encoded frame stored in is used as a reference frame, block matching is performed between the encoded frame and the reference frame, a motion amount called a motion vector is obtained, and motion compensation is performed from the reference frame based on this motion amount. To generate a predicted image signal. The predicted image signal generated in this way is supplied to the residual signal generation unit 102. Coding information such as a motion vector obtained by the predicted image generation unit 108 is stored in the coding information storage memory 113 as necessary. Furthermore, when a plurality of prediction modes can be selected, the prediction image generation unit 108 evaluates the distortion amount between the generated prediction image signal and the original image signal, and thereby selects an optimum prediction mode. The prediction image signal generated by the prediction in the determined prediction mode is selected and supplied to the residual signal generation unit 102. When the prediction mode is intra prediction, the intra prediction mode is selected as the encoded information storage memory. 113 and the first encoded bit string generator. Detailed description of the intra prediction mode will be described later.

The activity calculation unit 109 calculates an activity that is a coefficient indicating the complexity and smoothness of the image of the encoding target block supplied from the image memory 101, and supplies it to the quantization parameter calculation unit 110. The detailed configuration and operation of the activity calculation unit 109 will be described in an embodiment described later.

The quantization parameter calculation unit 110 calculates the quantization parameter of the block to be encoded based on the activity calculated by the activity calculation unit 109, and supplies the quantization parameter to the differential quantization parameter generation unit 111 and the encoded information storage memory 113. . The detailed configuration and operation of the quantization parameter calculation unit 110 will be described in an embodiment described later.

The difference quantization parameter generation unit 111 performs subtraction on the quantization parameter calculated by the quantization parameter calculation unit 110 and the prediction quantization parameter derived by the prediction quantization parameter derivation unit 114 to obtain the difference quantum. The calculation parameter is calculated and supplied to the first encoded bit string generation unit 112.

The first encoded bit sequence generation unit 112 generates a first encoded bit sequence by encoding the differential quantization parameter calculated by the differential quantization parameter generation unit 111 according to a specified syntax rule, and generates an encoded bit sequence multiplexed To the conversion unit 115.

The encoding information storage memory 113 stores the quantization parameter of the block that has been encoded. Although the connection is not shown in FIG. 1, the encoding information such as the prediction mode and the motion vector generated by the prediction image generation unit 108 is also used as information necessary for encoding the next encoding target block. Store. Furthermore, encoded information generated in units of pictures and slices is also stored as necessary.

The predictive quantization parameter deriving unit 114 derives a predictive quantization parameter using quantization parameters and coding information of already-encoded blocks that are close to the periphery of the block to be encoded, and generates a differential quantization parameter. Supplied to the unit 111. Predictive quantization parameter derivation unit 11
The detailed configuration and operation of 4 will be described in an embodiment described later.

The encoded bit string multiplexing unit 115 multiplexes the first encoded bit string and the second encoded bit string in accordance with a specified syntax rule, and outputs a bit stream.

2 shows a moving picture decoding apparatus 20 according to an embodiment corresponding to the moving picture encoding apparatus 100 of FIG.
FIG. The moving picture decoding apparatus 200 according to the embodiment includes a bit string separation unit 201, a first encoded bit string decoding unit 202, a quantization parameter generation unit 203, an encoded information storage memory 204, a predicted quantization parameter derivation unit 205, and a second code. Bit string decoding unit 2
06, inverse quantization / inverse orthogonal transform unit 207, decoded image signal superimposing unit 208, predicted image generating unit 20
9 and a decoded image memory 210. As in the moving picture encoding apparatus 100 of FIG. 1, the thick solid line arrows connecting the blocks represent picture image signals, and the thin solid line arrows represent the flow of parameter signals for controlling encoding.

The decoding process of the moving picture decoding apparatus 200 in FIG. 2 corresponds to the decoding process provided in the moving picture encoding apparatus 100 in FIG. 1, and therefore the inverse quantization / inverse orthogonal transform unit in FIG. 2. 207
1, the decoded image signal superimposing unit 208, the predicted image generating unit 209, the decoded image memory 210, and the encoded information storage memory 204 are configured by the inverse quantization / inverse orthogonal transform unit 105 of the moving image encoding apparatus 100 in FIG. Decoded image signal superimposing unit 106, predicted image generating unit 108, and decoded image memory 107
And a function corresponding to each component of the encoded information storage memory 113.

The bit stream supplied to the bit string separation unit 201 is separated according to a rule of a prescribed syntax, and the separated encoded bit string is supplied to the first encoded bit string decoding unit 202 and the second encoded bit string decoding unit 206.

The first encoded bit string decoding unit 202 decodes the supplied encoded bit string, outputs encoding information related to the prediction mode, motion vector, differential quantization parameter, and the like, and converts the differential quantization parameter to the quantization parameter generation unit 203. And the encoded information is stored in the encoded information storage memory 204.

The quantization parameter generation unit 203 adds the difference quantization parameter supplied from the first encoded bit string decoding unit 202 and the quantization parameter derived by the prediction quantization parameter deriving unit 205 to calculate the quantization parameter. And supplied to the inverse quantization / inverse orthogonal transform unit 207 and the encoded information storage memory 204.

The encoded information storage memory 113 stores the quantization parameter of the block that has been decoded. Furthermore, not only the block unit encoded information decoded by the first encoded bit string decoding unit 202 but also the encoded information generated in units of pictures and slices are stored as necessary. Although not shown in FIG. 2, the encoded information such as the decoded prediction mode and motion vector is supplied to the prediction image generation unit 209.

The predictive quantization parameter deriving unit 205 derives a predictive quantization parameter using the quantization parameter and encoding information of a block that has already been decoded near the decoding target block,
This is supplied to the quantization parameter generation unit 203. The predictive quantization parameter deriving unit 205 has a function equivalent to that of the predictive quantization parameter deriving unit 114 of the video encoding device 100, and a detailed configuration and operation will be described in an embodiment described later.

The second encoded bit string decoding unit 206 decodes the supplied encoded bit string to perform orthogonal transform /
The quantized residual signal is calculated, and the orthogonal transform / quantized residual signal is supplied to the inverse quantization / inverse orthogonal transform unit 207.

The inverse quantization / inverse orthogonal transform unit 207 generates a quantization signal generated by the quantization parameter generation unit 203 for the orthogonal transform / quantized residual signal decoded by the second encoded bit string decoding unit 206. Using the parameters, inverse orthogonal transform and inverse quantization are performed to obtain a residual signal subjected to inverse orthogonal transform and inverse quantization.

The decoded image signal superimposing unit 208 superimposes the predicted image signal generated by the predicted image generating unit 209 and the residual signal subjected to inverse orthogonal transform / inverse quantization by the inverse quantization / inverse orthogonal transform unit 207. The decoded image signal is generated, output, and stored in the decoded image memory 210.
When stored in the decoded image memory 210, the decoded image may be stored in the decoded image memory 210 after filtering processing for reducing block distortion or the like due to encoding is performed on the decoded image.

The predicted image generation unit 209 uses the decoded image memory 210 based on the encoded information such as the prediction mode and motion vector decoded by the second encoded bit string decoding unit 206 and the encoded information from the encoded information storage memory 204. A predicted image signal is generated from the decoded image signal supplied from, and supplied to the decoded image signal superimposing unit 208.

Next, various parts 120 surrounded by a thick dotted line in the moving picture coding apparatus 100, in particular, a predictive quantization parameter deriving part 114, and various parts 220 surrounded by a thick dotted line in the moving picture decoding apparatus 200, In particular, details of a method for deriving a predictive quantization parameter that is commonly performed by the predictive quantization parameter deriving unit 205 will be described.

First, the operation of each section of the sections 120 surrounded by the thick dotted line in the moving picture coding apparatus 100 according to the present embodiment will be described. The units 120 use a pixel block of a predetermined pixel size unit supplied from the image memory 101 as a coding block, and determine a quantization parameter for quantizing the block. The quantization parameter is mainly determined by code amount control and adaptive quantization algorithms. First, an adaptive quantization method in the activity calculation unit 109 will be described.

In the activity calculation unit 109, since human visual characteristics are generally sensitive to low-frequency components with few edges, a more complex portion of a pattern that is quantized more finely on a flat part that is visually conspicuous and is relatively inconspicuous. The activity expressing the complexity and smoothness of the image is calculated for each predetermined block so as to quantize more roughly.

As an example of the activity, there is a calculation based on a variance value of pixels in a coding block described in MPEG-2 TestModel 5 (TM5). The variance value is a value indicating the degree of dispersion from the average of the pixels constituting the image in the block. The flatter the image in the block (the smaller the luminance change), the smaller the value, and the more complex the image (the luminance change is). The larger the value, the larger the value, so use it as a block activity. When the pixel value in the block is represented by p (x, y), the activity act of the block is calculated by the following equation.

Here, BLK is the total number of pixels in the coding block, and p_mean is the average value of the pixels in the block.

Further, the present invention is not limited to the above variance, and the absolute value of the difference between the pixels in the coding block and the pixels adjacent in the horizontal direction and the vertical direction may be taken, and the sum in the block may be taken. Even in this case, it is small when the image is flat, and a large value is obtained in a complicated picture portion having many edges, and can be used as an activity. It is calculated by the following formula.

The activity act calculated in this way is supplied to the quantization parameter calculation unit 110.

Next, code amount control will be described. The moving image coding apparatus 100 according to the present embodiment does not particularly include a unit that realizes the code amount control. However, in the code amount control, the quantization parameter of the coding block is determined based on the generated code amount. It is assumed that the function is included in the parameter calculation unit 110.

Code amount control is aimed at adjusting the generated code amount of a predetermined unit such as a frame to the vicinity of the target code amount, and when it is determined that the generated code amount of the encoded block is larger than the target code amount, If relatively coarse quantization is applied to the block to be encoded, and it is determined that the generated code amount of the encoded block is less than the target code amount, relatively fine quantization is applied to the block to be encoded thereafter. To do.

  A specific code amount control algorithm will be described with reference to FIG.

First, a target code amount (T) is determined for each frame. In general, T is determined so that I picture> P picture> reference B picture> non-reference B picture. For example, when the target bit rate of a moving image is 5 Mbps, there are 1 I picture, 3 P pictures, 11 reference B pictures, and 15 non-reference B pictures per second, Assuming that the target code amount is Ti, Tp, Tbr, and Tb, Ti: Tp: Tbr: Tb = 4: 3:
When it is desired to control the target code amount so that the ratio is 2: 1, Ti = 400 kbit, Tp =
300 kbit, Tbr = 200 kbit, Tb = 100 kbit. However, the allocated code amount for each picture type does not affect the essence of the present invention.

となる。ここで、ｊは符号化ブロックの符号化処理順カウント番号である。Ｄ（０）は目
標符号量差分の初期値である。 Next, code amount control within a frame will be described. If the number of blocks, which are units for determining the quantization parameter, is N, the generated code amount is B, and the difference bit from the target code amount is D,

It becomes. Here, j is the encoding processing order count number of the encoding block. D (0) is an initial value of the target code amount difference.

The quantization parameter bQP by code amount control is determined as follows.

Here, r is a proportional coefficient for converting the target code amount difference into a quantization parameter. The proportional coefficient r is determined according to the available quantization parameter.

The quantization parameter calculation unit 110 changes the quantization parameter of the coding block calculated by the code amount control using the activity act calculated by the activity calculation unit 109 for each coding block. In the following, since the calculation is performed for each coding block, the coding processing order count number of the quantization parameter by the code amount control is deleted and represented by bQP.

The quantization parameter calculation unit 110 calculates an optimal quantization parameter QP for the coding block based on the activity act calculated by the activity calculation unit 109.

The quantization parameter calculation unit 110 records the average activity in the frame encoded immediately before as avg_act, and normalizes the encoded block activity Nact.
Is calculated by the following equation.

Here, the coefficient 2 in the above equation is a value representing the dynamic range of the quantization parameter.
A normalized activity Nact taking a range of 5 to 2.0 is calculated.

Note that avg_act may be calculated in advance for all blocks in the frame before the encoding process, and the average value may be avg_act. Furthermore, avc_a
ct may be stored in the encoded information storage memory 113, and the quantization parameter calculation unit 110 may acquire avg_act from the encoded information storage memory 113 as necessary.

The calculated normalization activity Nact is multiplied by the reference quantization parameter bQP as in the following equation to obtain the quantization parameter QP of the coding block.

Note that bQP is a quantization parameter for each block calculated by code amount control as described above, but may be a quantization parameter representative of a frame or slice including a coding block as a fixed value. Moreover, the average quantization parameter of the frame encoded immediately before may be used, and the calculation method is not particularly limited in the present embodiment.

The quantization parameter of the coding block calculated in this way is the coding information storage memory 11.
3 and the difference quantization parameter generation unit 111.

The encoded information storage memory 113 not only stores the quantization parameters calculated by the quantization parameter calculation unit 110 and the quantization parameters of past encoded blocks that have already been encoded, Encoding information such as a motion vector to be encoded and a prediction mode is also stored, and each unit acquires the encoding information as necessary.

The predicted quantization parameter derivation unit 114 uses the quantization parameter of the already-encoded neighboring block around the coding block and other coding information from the coding information storage memory 113 to calculate the quantization parameter of the coding block. Predictive quantization parameters for efficient coding and transmission are derived.

In order to efficiently encode and transmit the quantization parameter, rather than encoding the quantization parameter as it is, the difference (difference quantization parameter) with the quantization parameter of the already coded block is taken, It is more efficient to encode and transmit the differential quantization parameter. From the viewpoint of code amount control, if the quantization parameter of the immediately previous coded block in the coding processing order is the predicted quantization parameter, the value of the differential quantization parameter to be transmitted is small, and the code amount is small. On the other hand, from the viewpoint of adaptive quantization, since the coded block and the neighboring neighboring blocks are close to each other, the same or similar pattern is often obtained. If the quantization parameter of the adjacent block is set as the predictive quantization parameter, the value of the differential quantization parameter to be transmitted becomes small and the code amount becomes small. Therefore, H.H. In H.264, as shown in FIG. 4, the unit for transmitting the quantization parameter is fixed in a macro block (16 × 16 pixel group), and the left adjacent data encoded before or immediately before the encoded block in the raster scan order. A method is adopted in which the quantization parameter of the block adjacent to the block is used as the prediction quantization parameter, the difference between the quantization parameter of the coding block and the prediction quantization parameter is taken, and the difference quantization parameter is encoded and transmitted. That is, H.I. H.264 is optimized for prediction of quantization parameters assuming code amount control. However, H. Since H.264 does not perform hierarchical tree coding, which will be described later, since the immediately preceding block is the left block except for the left end of the image, the quantization parameter of the adjacent block is used as the prediction quantization parameter, and adaptive quantization is performed. It can be said that it is almost optimized for forecasting. Therefore, H.H. When the unit for transmitting the quantization parameter is fixed as in H.264 and the hierarchical tree coding is not performed, it can be said that the immediately preceding coded block is optimal for prediction of the quantization parameter.

However, when performing hierarchical tree coding, As in the case of H.264, when the quantization parameter of the immediately preceding block is used as the prediction quantization parameter, it is optimized for code amount control. However, when transmitting the quantization parameter using adaptive quantization, In other words, there arises a problem that the code amount of the differential quantization parameter increases.

Here, hierarchical tree coding will be described. Hierarchical tree coding here refers to de representing each coding unit in tree block units (here, 64 × 64 blocks).
pth is determined, and encoding is performed in units of encoded blocks with the determined depth. As a result, it is possible to determine the optimum depth depending on the definition of the image and perform the encoding, thereby greatly improving the encoding efficiency.

FIG. 5 shows the encoding processing order of the hierarchical tree encoding structure. As shown in the upper diagram of FIG.
Divide the screen equally into square units of the same size. This unit is called a tree block, and is a basic unit of address management for specifying an encoding / decoding block in an image. In order to optimize the encoding process according to the texture etc. in the image, the tree block,
If necessary, the tree block can be divided into four hierarchically to make a block having a small block size. Such a hierarchical block structure divided into small blocks is called a tree block structure, and the divided blocks are coded blocks (CU: C
is called uniting unit) and is a basic unit of processing when encoding and decoding. FIG.
The lower figure shows an example in which, among the CUs obtained by dividing the tree block into four parts, three CUs other than the lower left part are further divided into four parts. In this embodiment, it is assumed that the quantization parameter is set for each CU. The tree block is also the maximum size coding block.

In such hierarchical tree coding, the coding order is H.264 in FIG. Since this is different from the raster scan order (left to right) such as H.264, the quantization parameter may not be equal between the immediately preceding coded block and the left neighboring block. For example, as an example of hierarchical tree coding, as shown in FIG. 6, the upper left coding block (the shaded rectangle in FIG. 6) in the tree block to be coded is a tree block adjacent to the left. The quantization parameter of the lower right encoded block (the gray rectangle in FIG. 6) encoded last is used for prediction. Also, as shown in FIG. 7, the lower left coding block (shaded rectangle in FIG. 7) in the tree block to be coded is divided in the same tree block and coded immediately before. The quantization parameter of the block (the gray rectangle in FIG. 7) is used for prediction. Therefore, even if the prediction optimized for the code amount control can be performed only by predicting the quantization parameter from the previous coded block, adaptive quantization is used to separate the distance between the blocks by division. Therefore, the code amount of the differential quantization parameter increases, and the coding efficiency is reduced.

H. For example, in the case of the example shown in FIG. 8, since the picture pattern of the encoded block and the adjacent block on the left are different, the quantization parameter of the block on the left adjacent to the left like 264 is a predictive quantization parameter. The difference quantization parameter also becomes a large value, the amount of generated codes increases, and there is a possibility that efficient encoding and transmission cannot be performed.

As a solution to this, a method is conceivable in which the predicted quantization parameter is not selected from the left neighboring block, and the quantization parameter of the already-encoded neighboring block is used as the predicted quantization parameter.

However, when the quantization parameter is predicted from the neighboring block above the tree block boundary, the quantization parameter calculated at a considerably earlier time than the coding block is considered in consideration of the calculation of the quantization parameter by the code amount control. Therefore, as shown in FIG. 9, the processing order i of the upper neighboring block with respect to the processing order j of the coding block is i in the coding processing order even if it is close as a block in the picture. Since << j, from the viewpoint of code amount control, it cannot be said that the quantization parameter of the coding block is highly correlated with the quantization parameter of the upper neighboring block.

Furthermore, when performing parallel processing for each tree block slice to speed up the decoding process, the quantization parameter of the adjacent block above the tree block boundary cannot be used for prediction. By referring to the neighboring block above the tree block boundary, there is a possibility that efficient coding and transmission cannot be performed.

Therefore, the predictive quantization parameter deriving unit 114 according to the embodiment of the present invention does not use the neighboring blocks of the tree block adjacent on the coding block for the prediction of the quantization parameter, and the surrounding already coded The optimal predictive quantization parameter is derived from this block, and the efficiency of the generated code amount of the differential quantization parameter is improved.

FIG. 10 is a diagram showing a detailed configuration of the predicted quantization parameter deriving unit 114. The predicted quantization parameter derivation unit 114 includes a switch 301, a memory 302, a first predicted quantization parameter derivation unit 303, and a second predicted quantization parameter derivation unit 304.

Encoding information such as quantization parameters of encoded neighboring blocks and encoding block prediction modes around the encoded block supplied from the encoded information storage memory 113 are stored in the memory 3.
02 is temporarily stored, and the prediction mode of the encoded block is supplied to the switch 301.

The switch 301 switches a derivation unit that performs prediction of the quantization parameter according to the prediction mode. When the prediction mode is inter prediction, the first prediction quantization parameter deriving unit 303 derives the prediction quantization parameter, and when the prediction mode is intra prediction, the second prediction quantization parameter deriving unit 304 derives the prediction quantization parameter.

The first predictive quantization parameter deriving unit 303 derives a predictive quantization parameter from the memory 302 by relative comparison of quantization parameters of neighboring blocks that have already been encoded around the encoded block.

On the other hand, the second predictive quantization parameter deriving unit 304 selects from the memory 302 the quantization parameters of the already-encoded neighboring blocks around the encoded block and intra prediction as the prediction mode by the predictive image generation unit 108. Intra prediction mode (hereinafter referred to as intraPredMode) indicating the direction of the pixel of the adjacent block referred to in intra prediction
Thus, a predictive quantization parameter is derived.

Here, details of intra prediction will be described. In the intra prediction, the pixel value of the encoded block is predicted from the pixel values of the surrounding decoded blocks in the same screen. In the video encoding device 100 and the video decoding device 200 of the present invention, 34 intra prediction modes are selected and intra prediction is performed.

FIG. 11 is a diagram for explaining the value of the intra prediction mode and the prediction direction defined in the present embodiment. A solid arrow indicates a direction to be referred to in the intra prediction, and the intra prediction is performed from the end point of the arrow toward the start point. The number indicates the value of the intra prediction mode. The intra prediction mode is a vertical prediction (intra prediction mode intraP) for predicting in the vertical direction from the decoded block above.
redMode = 0), horizontal prediction for predicting in the horizontal direction from the left decoded block (intra prediction mode intraPredMode = 1), and average value prediction for predicting by calculating an average value from surrounding decoded blocks (intra prediction) Mode intraPredM
ode = 2), in addition to the average value prediction (intraPredMode = 3) that predicts in the lower right direction at an angle of 45 degrees obliquely from the surrounding decoded blocks, in an oblique direction at various angles from the surrounding decoded blocks 30 angle predictions to be predicted (intra prediction mode intraP
redMode = 4, 5,..., 33) is defined.

The intra prediction mode is prepared for each of the luminance signal and the color difference signal, and the intra prediction mode for the luminance signal is defined as the intra luminance prediction mode, and the intra prediction mode for the color difference signal is defined as the intra color difference prediction mode. In the encoding and decoding of the intra luminance prediction mode, using the correlation with the intra luminance prediction mode of the surrounding block, when it is determined that the encoding side can predict from the intra luminance prediction mode of the surrounding block, When it is determined that it is better to set a different value for the intra luminance prediction mode than for predicting from the intra luminance prediction mode of the surrounding blocks by transmitting information specifying the block to be referred to, the intra luminance prediction mode is further increased. A mechanism for encoding or decoding the value of is used. By predicting the intra luminance prediction mode of the block to be encoded / decoded from the intra luminance prediction modes of the surrounding blocks,
The amount of code to be transmitted can be reduced. On the other hand, in the coding and decoding of the intra chrominance prediction mode, the correlation between the prediction block of the luminance signal at the same position as the prediction block of the chrominance signal and the intra luminance prediction mode of the prediction block is used. If it is determined that the prediction can be made, the value of the intra color difference prediction mode is predicted from the value of the intra luminance prediction mode, and it is determined that it is better to set a unique value for the intra color difference prediction mode than to predict from the intra luminance prediction mode. In such a case, a mechanism for encoding or decoding the value of the intra color difference prediction mode is used. By predicting the intra color difference prediction mode from the intra luminance prediction mode, the amount of code to be transmitted can be reduced. In the present embodiment, unless otherwise specified, the intra luminance prediction mode is set to the intra prediction mode.

The predicted image generation unit 108 selects the intra prediction mode with the highest coding efficiency from the 34 intra prediction modes, and when the intra prediction mode is selected as the prediction mode, the intra prediction mode is selected together with the prediction mode. This is supplied to the predictive quantization parameter deriving unit 114 via the encoded information storage memory 113.

The second predicted quantization parameter deriving unit 304 derives a predicted quantization parameter based on this intra prediction mode. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

The differential quantization parameter generation unit 111 subtracts the prediction quantization parameter derived by the prediction quantization parameter deriving unit 114 from the quantization parameter of the coding block calculated by the quantization parameter calculation unit 110. And calculate a differential quantization parameter. Since the predictive quantization parameter is derived from the neighboring neighboring blocks that have already been decoded at the time of decoding as well as at the time of encoding, there is no contradiction in encoding and decoding by making the differential encoding parameter an encoding target, It is possible to reduce the code amount of the quantization parameter. The calculated difference quantization parameter is supplied to the first encoded bit string generation unit 112.

The first encoded bit string generation unit 112 generates a first encoded bit string by entropy encoding the differential quantization parameter calculated by the differential quantization parameter generation unit 111 according to a specified syntax rule. FIG. 12 shows an example of a coding conversion table used for entropy coding of the differential quantization parameter. This is a table called signed exponential Golomb coding, and a shorter code length is given as the absolute value of the differential quantization parameter is smaller. In general, when an image is divided into blocks, similar images are obtained in adjacent blocks, so that the activity is a close value, and the calculated quantization parameter of the block is also a close value. Therefore, the occurrence frequency of the difference quantization parameter tends to be the highest at 0 and lower as the absolute value becomes larger, and the table in FIG. Assigned a length. If the predicted quantization parameter is predicted with a value close to the quantization parameter of the coding block, a difference quantization parameter close to 0 is calculated, and the generated code amount can be suppressed. The first encoded bit string generation unit 112 extracts a code bit string corresponding to the differential quantization parameter from the table of FIG. 12, and supplies the code bit string to the encoded bit string multiplexing unit 115.

The operation of each unit of the units 220 surrounded by the thick dotted line in the video decoding device 200 corresponding to the video encoding device 100 of the above-described embodiment will be described.

In the units 220, the differential quantization parameter first decoded by the first encoded bit string decoding unit 202 is supplied to the quantization parameter generation unit 203. Also, encoded information other than the differential quantization parameter is stored in the encoded information storage memory 204 as necessary.

The quantization parameter generation unit 203 adds the differential quantization parameter supplied from the first encoded bit string decoding unit 202 and the quantization parameter derived by the prediction quantization parameter deriving unit 205 to obtain the quantization of the decoded block. The quantization parameter is calculated and supplied to the inverse quantization / inverse orthogonal transform unit 207 and the encoded information storage memory 204.

The encoded information storage memory 204 stores the quantization parameter of the block that has been decoded. Furthermore, not only the block unit encoded information decoded by the first encoded bit string decoding unit 202 but also the encoded information generated in units of pictures and slices are stored as necessary.

The predicted quantization parameter deriving unit 205 derives a predicted quantization parameter using the quantization parameters and coding information of already decoded blocks that are close to the periphery of the decoded block, and supplies them to the quantization parameter generating unit 203. . The quantization parameter calculated by the quantization parameter generation unit 203 is stored in the encoding information storage memory 204. When the predicted quantization parameter of the next decoding block is derived, the decoded neighboring blocks located around the decoding block are stored. The quantization parameter of the adjacent block is obtained from the encoded information storage memory 204. The quantization parameter of the decoded adjacent block obtained in this way is the same as the quantization parameter acquired from the encoded information storage memory 113 by the prediction quantization parameter deriving unit 114 of the video encoding device 100. Since the predictive quantization parameter deriving unit 205 has the same function as the predictive quantization parameter deriving unit 114 of the video encoding device 100, the quantization parameter of the adjacent block supplied from the encoded information storage memory 204 is If they are the same, the same predictive quantization parameter as that at the time of encoding is derived.

The predictive quantization parameter deriving unit 205 performs the same processing except that the encoded adjacent block is changed to the decoded adjacent block, and thus description of the quantization parameter prediction is omitted.

Thus, the predicted quantization parameter derived on the encoding side is derived on the decoding side without any contradiction.

In the present embodiment, when predictive quantization parameters are derived, the adjacent blocks referred to by the predictive quantization parameter deriving unit 114 of the video encoding device 100 are already encoded blocks, and the video decoding device 200 The adjacent block referred to by the predicted quantization parameter deriving unit 205 is a decoded block. The encoded block referenced on the encoding side is
This block is locally decoded for the next encoding within the encoding, and is the same as the decoded block referenced on the decoding side. Therefore, the functions of the predictive quantization parameter deriving units 114 and 205 are also common, and the predictive quantization parameters derived by each are the same. In the following embodiments, the derivation of the prediction quantization parameter will be described on the encoding side as a common function without being divided between encoding and decoding.

Hereinafter, the details of the method for deriving the predicted quantization parameter that is commonly performed by the predicted quantization parameter deriving units 114 and 205 will be described.

[Example 1]
A detailed operation of the first predictive quantization parameter deriving unit 303 in the first embodiment will be described. In the first embodiment, when the coding block to be coded is close to the upper tree block, the quantization parameter of the coded block in the upper tree block that is considerably past in the coding order is used for prediction. Although it is prohibited, the quantization parameter of the coded block of the tree block adjacent to the left which is past in the coding order but not as old as the upper tree block is used for the prediction.

As shown in FIG. 13, encoding is performed in the order of raster scanning from the upper left to the lower right of the screen in units of tree blocks. If the encoding target tree block is represented by a hatched rectangle in FIG. 13, the encoded tree block is represented by a gray portion in FIG. Since hierarchical tree coding is performed in the tree block according to the coding conditions, the coding block is divided into a size equal to or smaller than the tree block. Although it is close to the encoded block inside the tree block, it is far away in the encoding processing order. Therefore, since the quantization parameter calculated by the code amount control is calculated in the order of encoding processing, the quantization parameter of the encoding block is close to the quantization parameter of the encoded block in the upper tree block. I can't say. Therefore, in the first embodiment, the upper tree block is not used for the prediction of the quantization parameter, but only the left tree block in the encoding processing order is used.

Further, as shown in FIG. 14, when the encoding block in the tree block is a hatched rectangle in FIG. 14, the solid thin line represents the encoding processing order, and the blocks encoded up to the encoding block are shown in FIG. 14 is represented by a gray portion. Within the same tree block, the coding process order of the coded block and the coded block is not separated, and the same or similar pattern is often obtained. It is useful to use parameters for prediction. In the first embodiment, priority is given to using an adjacent encoded block for prediction rather than an encoded block close in the encoding processing order.

FIG. 15 shows the direction of the encoded block referred to by each encoded block in the divided tree block by a thick arrow. The solid thin line in FIG. 15 represents the encoding process order, and the encoded block gives priority to the adjacent encoded block over the encoded block close in the encoding process order. Since BLK0 and BLK1 located at the upper end of the tree block in FIG. 15 are in contact with the upper tree block, the quantization parameter of the encoded block adjacent to the upper side is not used for prediction, and is adjacent to the left. Only the quantization parameters of the encoded block are used. Since BLK2 and BLK3 have encoded blocks adjacent to each other in the same tree block, the quantization parameter of the upper encoded block and the quantization parameter of the left encoded block are used for prediction.

FIG. 16 shows the arrangement of the encoded blocks defined in the present embodiment and the already encoded blocks adjacent to the surroundings. In this embodiment, the size of each block is described as the same for convenience of explanation. However, even when optimal motion prediction is performed by changing the block size by, for example, motion prediction, the upper left of the encoded block This can be realized by selecting a block close to the periphery of the above point as a reference.

A symbol QPx (x = L, A, AL) shown in FIG. 16 represents a quantization parameter of a neighboring block that has already been encoded. The first predictive quantization parameter deriving unit 303 determines the predictive quantization parameter based on the presence / absence of the quantization parameter of the left and upper neighboring blocks shown in FIG.

The operation of the first predictive quantization parameter deriving unit 303 will be described. FIG. 17 shows the first embodiment.
6 is a flowchart showing the operation of the first predictive quantization parameter deriving unit 303 in FIG.

First, position information of a coding block to be coded is acquired (S100). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. It is then determined whether the encoded block is close to the upper tree block (
S101).

If the encoded block is close to the upper tree block (Yes in S101), that is, if the encoded block is located at the upper end of the tree block, the upper adjacent block is included in the upper tree block. Therefore, the tree block boundary is crossed, so the upper neighboring block is not used for the quantization parameter prediction. Here, focusing on the fact that the quantization parameter always takes a positive value, when the upper neighboring block is not used, the quantization parameter QPA of the upper neighboring block is set to 0 (S102).

On the other hand, when the encoded block is not close to the upper tree block (No in S101)
That is, when the upper adjacent block is located in the same tree block as the encoded block, the encoded data that is close to the encoded block from the memory 302 is determined from the reference position information at the upper left of the encoded block. The quantization parameter QPA of the completed block is acquired (S103).
At this time, the encoded information storage memory 113 is based on the reference position information at the upper left of the encoded block.
, The quantization parameter of the corresponding surrounding encoded block is supplied to the memory 302 in the prediction quantization parameter deriving unit 114, and the memory 30 is stored.
2. Quantization parameters of neighboring blocks above 2 are acquired.

Subsequently, it is determined whether there is an encoded block adjacent to the left of the encoded block (S104). When there is an adjacent block on the left (Yes in S104), the memory 302
To obtain the quantization parameter QPL of the encoded block adjacent to the left of the encoded block (S105). If the left adjacent block does not exist (No in S104), the quantization parameter QPL of the left adjacent block is set to 0 (S106).

Next, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S107). When the quantization parameters of the left and upper neighboring blocks are both positive (S107)
Yes), since both the left and upper neighboring blocks exist, the average value of the quantization parameters of the left and upper neighboring blocks is set as the predicted quantization parameter (S111). On the other hand, if the quantization parameter of the left and upper neighboring blocks is not positive (No in S107), that is, the quantization parameter of at least one of the left or upper neighboring blocks is 0, and at least either the left or upper one One of the adjacent blocks does not exist. In this case, S10
Proceed to step 8.

Next, it is determined whether the quantization parameters of the left and upper neighboring blocks are both 0 (S108). That is, when the quantization parameters of the left and upper neighboring blocks are both 0, both of them do not exist, so the quantization parameters of the left and upper neighboring blocks cannot be referred to as the prediction quantization parameters. Therefore, the quantization parameter (prevQP) of the block encoded before or immediately before the encoding block to be encoded is set as the predicted quantization parameter. If the upper left block of the image is an encoded block,
Further, since there is no block coded before or immediately before the coding block to be coded, the quantization parameter of the picture or slice is set as the prediction quantization parameter (S1).
09). When either the left or upper neighboring block exists, one positive quantization parameter is set as a predicted quantization parameter (S110). The predicted quantization parameter calculated in this way is supplied to the differential quantization parameter generation unit 111.

Also, the first predictive quantization parameter deriving unit 303 can determine the predictive quantization parameter based on the quantization parameters of the neighboring blocks on the left, top, and top left around the coding block shown in FIG. is there. The difference from the above-described method is that, based on the determination of the predicted quantization parameter, the quantization parameter of the left and upper neighboring blocks is weighted, and the derived value is used as the predicted quantization parameter.

FIG. 18 is a flowchart showing the operation of the first predictive quantization parameter deriving unit 303. The processing steps from S200 to S210 in the flowchart of FIG. 18 are the same as S100 to S110 in the flowchart of FIG. 17 described above, so the description thereof will be omitted, and whether the quantization parameters of the left and upper neighboring blocks are both positive or not. In the determination (S207), the description starts from the case where the quantization parameters of the left and upper neighboring blocks are both positive (Yes in S207). If the quantization parameters of the left and top neighboring blocks are both positive, there are both left and top neighboring blocks. At this time, since there is an adjacent block at the upper left, the storage area stored in the encoded information storage memory 113 is accessed from the reference position information at the upper left of the encoded block,
The quantization parameter QPAL of the corresponding upper left neighboring block is supplied to the predicted quantization parameter deriving unit 114 (S211).

Next, it is determined whether or not the quantization parameter QPL of the left neighboring block and the quantization parameter QPAL of the upper left neighboring block match (S212). When QPL and QPAL match, if the weighting factor FA of the quantization parameter of the upper neighboring block and the weighting factor FL of the quantization parameter of the left neighboring block, the quantization of the upper neighboring block is such that FA> FL. A large weight is set for the parameter (S213). For example, FA is set to 3 and FL is set to 1. In this case, since the arrangement of the quantization parameters shown in FIG. 8 can be considered as an example, it can be said that it is appropriate to set the weights of the quantization parameters of the upper neighboring blocks large. In addition, when QPA matches QPL and QPAL, the quantization parameters of all adjacent blocks are the same, so there is no problem. If QPL and QPAL do not match, S21
Proceeding to step 4, the coincidence determination between QPA and QPAL is performed (S214). If QPA and QPAL match, weighting is set larger for the quantization parameter of the adjacent block on the left so that FA <FL (S215). For example, FA is set to 1 and FL is set to 3. QPA and Q
If the PALs do not match, the FA and FL are set to the same weight, and the weights for the quantization parameters of the left and upper neighboring blocks are equalized (S216). In this case, since the quantization parameters of the left, upper, and upper left neighboring blocks are all different, it is not possible to perform sufficient condition determination to set a large weight for either QPL or QPA. Therefore, an average of QPL and QPA is used as a predictive quantization parameter, and an equal determination value is set. For example, FA is set to 2 and FL is set to 2. A predicted quantization parameter predQP is derived from the determined weighting coefficient and each quantization parameter by the following equation (S217).

Here, the denominator of the above formula is FA + FL, and 2 of the numerator is added for rounding off (FA
+ FL) / 2. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

Further, instead of the same determination of QPL and QPAL in S212 and QPA and QPAL in S214 in FIG. 18, the difference absolute value of the quantization parameter of the left and upper left neighboring blocks is ΔL, and the quantization of the upper and left neighboring blocks It is also possible to select the left or upper quantization parameter as the predicted quantization parameter based on the comparison between ΔL and ΔA, with the parameter difference absolute value as ΔA.

In the coded block and the neighboring blocks that have already been coded, ΔL and ΔA represent the absolute value of the difference between the quantization parameters of the left and upper left, and the upper and upper left neighboring blocks, respectively, and are expressed by the following equations.

When ΔA is larger than ΔL, the difference between QPA and QPAL is large, and the smoothness or complexity of the image differs between the upper and upper left neighboring blocks than between the left and upper left neighboring blocks. It is inferred that there is a large change. Therefore, in the coding block and the neighboring blocks that have already been coded, quantization is performed with the left two blocks (left and upper left neighboring blocks) and the right two blocks (coded block and upper neighboring blocks). Since a parameter difference is considered to occur, it is determined that the quantization parameter of the coding block is close to the quantization parameter of the neighboring block above the quantization parameter of the left neighboring block.

In the case of decoding processing, the prediction quantization parameter derivation unit replaces the code from 303 to 205 and the encoding information storage memory from 302 to 204, and the output destination of the prediction quantization parameter is changed from the difference quantization parameter generation unit 111 to the quantization parameter. By using the generation unit 203, equivalent processing is realized.

[Example 2]
The operation of the first predictive quantization parameter derivation units 303 and 205 in the second embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 303 to 205, the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. In the second embodiment, as in the first embodiment, the quantization parameters of the left and upper encoded blocks adjacent to the encoding block to be encoded are used for prediction. On the other hand, the difference from the first embodiment is that when the encoded block is close to the upper tree block, the quantization of the encoded block in the left tree block is performed when the encoded block is close to the left tree block. The use of parameters for prediction is prohibited. This is because the quantization parameter of the coding block is calculated based on the coding processing order of the coding control, so that the coding processing order is separated between the coding blocks between the tree blocks as compared to the inside of the tree block. Even if coding blocks are close to each other between the tree blocks, the quantization parameter of the coding block calculated by the code amount control is not necessarily a close value and may not be suitable as a predictive quantization parameter. is there. Therefore, in the second embodiment, when the encoding block to be encoded or decoded is close to the left or upper tree block, the quantization parameter of the encoded block in the left or upper tree block is predicted. Not used, but replaced with the quantization parameter of the block encoded before or immediately before the encoding block to be encoded in the encoding processing order.

FIG. 19 shows the direction of the encoded block referred to by each encoded block in the divided tree block by a thick arrow. The solid thin line in FIG. 19 represents the order of encoding processing, and the encoding block uses in principle the quantization parameters of the adjacent encoded blocks within the tree block including the encoding block. Since BLK0 located at the upper end of the tree block in FIG. 20 is in contact with the left and upper tree blocks, before the coding block to be coded,
Or, the quantization parameter of the block coded immediately before is replaced with the quantization parameter of the coded block adjacent to the left and the top and used for prediction. Since BLK1 is bordered by the upper tree block, the quantization parameter of the encoded block adjacent to the upper side is not used for prediction, and the block encoded before or immediately before the encoding block to be encoded is used. Is used for prediction together with the quantization parameter of the coded block adjacent to the left. Since BLK2 is in contact with the left tree block, the quantization parameter of the encoded block adjacent to the left is not used for prediction, and the block is encoded before or immediately before the encoding block to be encoded. Is used for prediction together with the quantization parameter of the coded block close to the top. BLK3 uses the quantization parameter of the upper encoded block and the quantization parameter of the left encoded block for prediction since the adjacent encoded blocks on the left and upper sides are inside the same tree block.

FIG. 20 is a flowchart illustrating the operation of the first predictive quantization parameter deriving unit 303 in the second embodiment.

First, position information of a coding block to be coded is acquired (S300). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. It is then determined whether the encoded block is close to the upper tree block (
S301). When the encoding block is close to the upper tree block (Yes in S301)
s), that is, when the coding block is located at the upper end of the tree block, since the upper neighboring block is included in the upper tree block, the tree block boundary is exceeded. The quantization parameter prevQP of the coding block coded before or just before the coding block to be coded is set to QPA without using the adjacent block of (S302).

On the other hand, when the coding block is not close to the upper tree block (No in S301)
That is, when the upper adjacent block is located in the same tree block as the encoded block, the quantization parameter QPA of the already encoded block adjacent to the encoded block is obtained from the memory 302. (S303). At this time, the storage area stored in the encoded information storage memory 113 is accessed based on the reference position information at the upper left of the encoded block, and the quantization parameter of the corresponding surrounding encoded block is the predicted quantization parameter. Deriving unit 114
Supplied to the internal memory 302, the quantization parameter of the upper neighboring block is acquired from the memory 302.

Subsequently, it is determined whether or not the encoded block is close to the left tree block (S
304). When the coding block is close to the left tree block (Yes in S304)
), I.e., if the coding block is located at the left end of the tree block, the left neighboring block is included in the left tree block, and thus the tree block boundary is exceeded. Without using adjacent blocks, before the encoding block to be encoded,
Or the quantization parameter prevQP of the coding block coded immediately before is set to QPL (S306).

On the other hand, when the coding block is not close to the left tree block (No in S304)
That is, if the left adjacent block is located in the same tree block as the encoded block, the quantization parameter QPL of the already encoded block adjacent to the left of the encoded block is acquired from the memory 302. (S306). Finally, the average value of the quantization parameters of the adjacent blocks on the left and upper is set as the predicted quantization parameter (S307). The predicted quantization parameter calculated in this way is supplied to the differential quantization parameter generation unit 111.

In the second embodiment, when the left and upper neighboring blocks exceed the tree block boundary, the quantization parameters of the coding blocks coded before or immediately before the coding block to be coded are encoded. Therefore, it is possible to reduce the determination process of the quantization parameter value as compared with the first embodiment.

[Example 3]
The operation of the first predictive quantization parameter derivation units 303 and 205 in the third embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 303 to 205, the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. The difference from the first embodiment is that, when the encoding block to be encoded or decoded is close to the upper tree block, the encoding in the left tree block is already performed when it is close to the left tree block. The point is to prohibit the use of block quantization parameters for prediction. That is, the quantization parameter of the encoded block that exceeds the tree block boundary is used for prediction because the first encoding block in the tree block is preceded by the encoding block to be encoded. Or, it is limited only when the quantization parameter of the block coded immediately before is used.

FIG. 21 shows the direction of the encoded block referred to by each encoded block in the divided tree block by a thick arrow. The solid thin line in FIG. 21 represents the encoding processing order, and the encoding block uses the quantization parameter of the encoded block that is close within the tree block including the encoding block.

Since BLK0 located at the upper end of the tree block in FIG. 21 is in contact with the left and upper tree blocks, only the quantization parameter of the block encoded before or immediately before the encoding block to be encoded is predicted. Used for. Since BLK1 is bounded by the upper tree block, it does not use the quantization parameter of the encoded block close to the top for prediction,
Only the quantization parameter of the encoded block close to the left is used for prediction. Since BLK2 is bounded by the left tree block, the quantization parameter of the encoded block adjacent to the left is not used for prediction, and only the quantization parameter of the encoded block adjacent to the left is used for prediction. . BLK3 uses the quantization parameter of the upper encoded block and the quantization parameter of the left encoded block for prediction since the adjacent encoded blocks on the left and upper sides are inside the same tree block.

FIG. 22 is a flowchart illustrating the operation of the first predictive quantization parameter deriving unit 303 in the third embodiment. S400 to S403 and S407 to S in the flowchart of FIG.
Steps up to 411 are the same as steps S100 to S103 and steps S107 to S111 in FIG. 17 of the first embodiment, and thus the description is omitted and S404 after it is determined whether or not the encoded block is close to the upper tree block. Only the differences from are described.

After the proximity determination between the encoded block and the upper tree block, it is determined whether or not the encoded block is adjacent to the left tree block (S404). If the encoded block is close to the left tree block (Yes in S404), that is, if the encoded block is located at the left end of the tree block, the left adjacent block is included in the left tree block. Therefore, since the tree block boundary is exceeded, the left adjacent block is not used for the quantization parameter prediction. Here, focusing on the fact that the quantization parameter always takes a positive value, when the left adjacent block is not used, the quantization parameter QPL of the left adjacent block is set to 0 (S
405). On the other hand, when the coding block is not close to the left tree block (S404).
No), that is, when the adjacent block on the left is located in the same tree block as the encoded block, the quantization parameter QPL of the already encoded block adjacent to the left of the encoded block from the memory 302 Is acquired (S406). A predicted quantization parameter is derived from the quantization parameters of the left and upper neighboring blocks acquired in this way, and the predicted quantization parameter is supplied to the differential quantization parameter generation unit 111.

[Example 4]
The operation of the first predictive quantization parameter deriving units 303 and 205 in the fourth embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 303 to 205, the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. In the fourth embodiment, when the encoding block to be encoded or decoded is close to the left or upper tree block, the quantization parameter of the encoded block in the left or upper tree block is used for prediction. Is prohibited. Furthermore, in principle, the quantization parameter of the encoded block adjacent to the left is used for the prediction, and if there is no encoded block adjacent to the left or located beyond the tree block boundary, the encoding is performed. The quantization parameter of the block encoded before or immediately before the target encoded block is used for prediction.

In FIG. 23, the direction of the encoded block referred to by each encoded block in the divided tree block is represented by a thick arrow. The solid thin line in FIG. 23 represents the encoding processing order, and the encoding block uses the quantization parameter of the encoded block adjacent to the left inside the tree block including the encoding block in principle for prediction.

Since BLK0 located at the upper end of the tree block in FIG. 23 is in contact with the left and upper tree blocks, the quantization parameter of the block encoded before or immediately before the encoding block to be encoded is predicted. Used for. BLK1 and BLK3 use the quantization parameter of the left encoded block for prediction because the encoded block adjacent to the left is inside the same tree block. Since BLK2 borders the left tree block,
The quantization parameter of the coded block adjacent to the left is not used for prediction, and the quantization parameter of the block coded before or just before the coding block to be coded is used for prediction.

FIG. 24 is a flowchart illustrating the operation of the first predictive quantization parameter deriving unit 303 in the fourth embodiment. First, the position information of the encoding block to be encoded is acquired (S50).
0). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. Next, it is determined whether or not the encoded block is close to the left tree block (S501). If the encoded block is close to the left tree block (Yes in S304), that is, if the encoded block is located at the left end of the tree block, the left adjacent block is included in the left tree block. For this reason, since the tree block boundary is exceeded, the left neighboring block is not used for the quantization parameter prediction, and the quantization parameter prev of the coding block coded before or just before the coding block to be coded is used.
QP is set as a predictive quantization parameter (S502). On the other hand, if the encoded block is not adjacent to the left tree block (No in S501), that is, if the left adjacent block is located in the same tree block as the encoded block, the code is read from the memory 302. A quantization parameter QPL of an already-encoded block adjacent to the left of the quantization block (
S503). At this time, the storage area stored in the encoded information storage memory 113 is accessed based on the reference position information at the upper left of the encoded block, and the quantization parameter of the corresponding surrounding encoded block is the predicted quantization parameter. The quantization parameter of the adjacent block on the left is acquired from the memory 302 inside the derivation unit 114, and set as the prediction quantization parameter. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

In the fourth embodiment, since the quantization parameter of the left neighboring block that has been encoded of the encoded block is used for the principle prediction, the determination process is simplified compared to the previous embodiments,
It is possible to reduce the circuit scale.

[Example 5]
The operation of the first predictive quantization parameter derivation units 303 and 205 in the fifth embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 303 to 205, the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. The fifth embodiment is a combination of the first and second embodiments. When the encoding block to be encoded or decoded is close to the left tree block, the quantization parameter of the encoded block of the tree block adjacent to the left is set. Allow use for prediction. If the coded block is close to the upper tree block, it is prohibited to use the quantization parameter of the coded block in the upper tree block for prediction, and the coded block in the upper tree block is used. Instead of the quantization parameter, the quantization parameter of the block coded before or just before the coding block to be coded is used for prediction.

FIG. 25 shows the direction of the encoded block referred to by each encoded block in the divided tree block by a thick arrow. The solid thin line in FIG. 25 represents the encoding process order, and the encoding block to be encoded gives priority to the adjacent encoded block over the encoded block close in the encoding process order.

Since BLK0 and BLK1 located at the upper end of the tree block in FIG. 25 are bounded by the upper tree block, the quantization parameter of the encoded block adjacent to the upper side is not used for prediction, and encoding is performed instead. The quantization parameter of the block coded before or immediately before the target coding block and the quantization parameter of the coded block adjacent to the left are used for prediction. Since BLK2 and BLK3 have encoded blocks adjacent to each other in the same tree block, the quantization parameter of the upper encoded block and the quantization parameter of the left encoded block are used for prediction.

A detailed operation of the first predictive quantization parameter deriving unit 303 in the fifth embodiment will be described. FIG. 26 is a flowchart illustrating the operation of the first predictive quantization parameter deriving unit 303 in the fifth embodiment.

First, position information of a coding block to be coded is acquired (S600). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. It is then determined whether the encoded block is close to the upper tree block (
S601). When the coded block is close to the upper tree block (Yes in S601)
s), that is, when the coding block is located at the upper end of the tree block, since the upper neighboring block is included in the upper tree block, the tree block boundary is exceeded. The quantization parameter prevQP of the coding block coded before or just before the coding block to be coded is set to QPA without using the adjacent block of (S602).

On the other hand, when the encoded block is not close to the upper tree block (No in S601)
That is, when the upper adjacent block is located in the same tree block as the encoded block, the quantization parameter QPA of the already encoded block adjacent to the encoded block is obtained from the memory 302. (S603). At this time, the storage area stored in the encoded information storage memory 113 is accessed based on the reference position information at the upper left of the encoded block, and the quantization parameter of the corresponding surrounding encoded block is the predicted quantization parameter. Deriving unit 114
It is supplied to the internal memory 302 and obtains the quantization parameter of the upper neighboring block from the memory 302.

Subsequently, it is determined whether there is an encoded block adjacent to the left of the encoded block (S604). When the adjacent block exists on the left (Yes in S604), the storage area stored in the encoded information storage memory 113 is accessed from the reference position information on the upper left of the encoded block, and the quantum of the corresponding left adjacent block is determined. The quantization parameter QPL is supplied to the predicted quantization parameter deriving unit 114 (S605). If the left neighboring block does not exist (No in S104), the quantization parameter QPL of the left neighboring block is set to 0 (S606).

Next, it is determined whether the quantization parameter of the left adjacent block is positive (S607).
When the quantization parameter of the left neighboring block is positive (Yes in S607), since the left neighboring block exists, the average value of the quantization parameters of the left and upper neighboring blocks is set as the predicted quantization parameter (S608). . On the other hand, when the quantization parameter of the left neighboring block is not positive (No in S607), that is, the quantization parameter of the left neighboring block is 0, and the left neighboring block does not exist. In this case, QPA is used as a predicted quantization parameter (
S609). The predicted quantization parameter thus calculated is the difference quantization parameter generation unit 1
11 is supplied.

[Example 6]
In the sixth embodiment, operations of the second prediction quantization parameter derivation units 304 and 205 will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 304 to 205, and the encoding information storage memory 30
By replacing the code from 2 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, equivalent processing is realized. The second predictive quantization parameter deriving unit 304 switches to the intra prediction mode indicating the pixel direction of the neighboring block referred to in the intra prediction when the prediction mode of the encoded block is selected by the predicted image generation unit 108 as the prediction mode of the encoded block. Based on this, a prediction quantization parameter is selected or derived.

FIG. 27 is a diagram illustrating a relationship between the prediction direction of the intra prediction mode and the left and upper neighboring blocks selected as the prediction quantization parameter when the prediction mode of the encoded block is intra prediction. In intra prediction, a prediction signal is generated from pixels in adjacent blocks in the prediction direction indicated by the intra prediction mode, and an intra prediction mode with high coding efficiency and low distortion is selected. Therefore, when the intra prediction mode selects the upper side, it is determined that the correlation with the upper neighboring block is high, and when the intra prediction mode selects the left side, the correlation with the left neighboring block is high. The higher the correlation between blocks, the more similar the pattern and the degree of complexity (activity) in the blocks, and the closer the quantization parameter calculated from the activity is. Therefore, in the sixth embodiment, in order to encode and transmit the quantization parameter of the coding block with high efficiency, a proximity block having high correlation with the coding block is predicted and determined in the intra prediction mode, and the determined proximity is determined. A block quantization parameter is selected as a predictive quantization parameter.

FIG. 28 is a flowchart illustrating the operation of the second predictive quantization parameter deriving unit 304 in the sixth embodiment. In the sixth embodiment, when the encoded block is close to the upper tree block, the quantization parameter of the encoded block in the upper tree block is prohibited from being used for prediction, but the tree close to the left is used. The quantization parameter of the encoded block of the block is used for prediction.

First, position information of a coding block to be coded is acquired (S700). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. It is then determined whether the encoded block is close to the upper tree block (
S701).

If the encoded block is close to the upper tree block (Yes in S701), that is, if the encoded block is located at the upper end of the tree block, the upper adjacent block is included in the upper tree block. Therefore, the tree block boundary is crossed, so the upper neighboring block is not used for the quantization parameter prediction. Here, focusing on the fact that the quantization parameter always takes a positive value, when the upper neighboring block is not used, the quantization parameter QPA of the upper neighboring block is set to 0 (S702).

On the other hand, when the encoded block is not close to the upper tree block (No in S701)
That is, when the upper adjacent block is located in the same tree block as the encoded block, the quantization parameter QPA of the already encoded block adjacent to the encoded block is obtained from the memory 302. (S703). At this time, the storage area stored in the encoded information storage memory 113 is accessed based on the reference position information at the upper left of the encoded block, and the quantization parameter of the corresponding surrounding encoded block is the predicted quantization parameter. Deriving unit 114
It is supplied to the internal memory 302, and the quantization parameter of the upper adjacent block is acquired from the memory 302.

Subsequently, it is determined whether there is an encoded block adjacent to the left of the encoded block (S704). When there is an adjacent block on the left (Yes in S704), the memory 302
To obtain the quantization parameter QPL of the encoded block adjacent to the left of the encoded block (S705). If the left adjacent block does not exist (No in S704), the quantization parameter QPL of the left adjacent block is set to 0 (S706).

Next, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S707). When the quantization parameters of the left and upper neighboring blocks are both positive (S707)
Yes), there are both left and top neighboring blocks. Therefore, it is possible to obtain the quantization parameter by referring to the adjacent block specified by the prediction direction indicated in the intra prediction mode. In this case, the prediction determination based on the prediction direction indicated in the intra prediction mode I do.

On the other hand, if the quantization parameter of the left and upper neighboring blocks is not positive (No in S707), that is, the quantization parameter of at least either the left or upper neighboring block is 0, and at least either the left or upper One of the adjacent blocks does not exist. Therefore, there may be no adjacent block specified in the prediction direction indicated in the intra prediction mode. In this case, the prediction block based on the prediction direction indicated in the intra prediction mode is not performed, and the adjacent block on the left or above Is determined based on the presence or absence of the data, that is, whether the quantization parameter of the adjacent block on the left or above is 0. First, it is determined whether the quantization parameter of the upper neighboring block is 0 (S710). When the quantization parameter of the upper neighboring block is not 0 (No in S710), the quantization parameter of the upper neighboring block is set as a predicted quantization parameter (S711). On the other hand, if it is 0 (Yes in S710), the quantization parameter of the above adjacent block cannot be referred to as the predicted quantization parameter, so S71
Proceed to determination 2. Next, it is determined whether or not the quantization parameter of the left adjacent block is 0 (S712). When the quantization parameter of the left neighboring block is not 0 (No in S712)
), The quantization parameter of the left adjacent block is set as the prediction quantization parameter (S713).
On the other hand, if it is 0 (Yes in S712), that is, the quantization parameters of the left and upper neighboring blocks are both 0, and neither exists, the quantization of the left and upper neighboring blocks as the predicted quantization parameter The parameter cannot be referenced. Therefore, the quantization parameter (prevQP) of the block encoded before or immediately before the encoding block to be encoded is set as the predicted quantization parameter. If the block at the upper left corner of the image is an encoded block, there is no block encoded before and immediately before the encoding block to be encoded, and the adjacent block on the left and upper, so a picture or slice Is set as a predicted quantization parameter (S714).

Returning to the determination in S707, when the quantization parameters of the left and upper neighboring blocks are both positive (Yes in S707), the intra prediction mode in of the encoded block is read from the memory 302.
Get traPredMode. Similarly to the acquisition of the quantization parameter of the surrounding encoded block, the intra prediction mode accesses the storage area stored in the encoded information storage memory 113 based on the reference position information at the upper left of the encoded block, The data is supplied to the memory 302 and acquired from the memory 302.

The intra prediction mode is a number assigned to the prediction direction shown in FIG.
In order to facilitate the following determination processing, the intra prediction mode is converted into an intra prediction direction (hereinafter referred to as intraPredDirec) using the conversion table shown in FIG.
The intra prediction direction is an average value prediction (intraP) in which intraPredDirec = 0 is calculated by calculating an average value from the surrounding decoded blocks in the intra prediction.
redMode = 2), and the other 1 to 33 are assigned from the upper right (intraPredMode = 6) to the lower left (intraPr) in the prediction direction of the intra prediction mode shown in FIG.
It is assigned so as to correspond to edMode = 9). With this conversion table, the acquired intra prediction mode is converted into the intra prediction direction (S708). The quantization parameter is predicted based on the converted intra prediction direction (S709). 30 is set as the threshold value of the prediction direction of the intra prediction mode shown in FIG. 30. Here, TH is set to 18 as the value of the intra prediction direction. If the quantization parameter is 18 or more, the quantization parameter of the left neighboring block is derived as the predicted quantization parameter.

The detailed operation of predicting the quantization parameter based on the intra prediction direction will be described with reference to FIG. First, it is determined whether or not the converted intra prediction direction is 0 (S720).
The intra prediction direction is 0 when intra prediction based on average value prediction is performed (
Yes in S720. In this case, either the left or the upper neighboring block is not preferentially selected, and the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks is set as the predicted quantization parameter (S721). On the other hand, when the intra prediction direction is not 0 (S7)
20 No), it is determined whether the intra prediction direction is less than 18 (S722). When the intra prediction direction is less than 18 (Yes in S722), the upper neighboring block is referred to as shown in FIG. 27, and the quantization parameter QPA of the upper neighboring block is used as the predicted quantization parameter (S723). ). On the other hand, when the intra prediction direction is 18 or more (S722)
No), the left neighboring block is referred to, and the quantization parameter QPL of the left neighboring block is set as the predicted quantization parameter (S724). The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

As described above, in the second predictive quantization parameter deriving unit 304 in the sixth embodiment, FIG.
Based on the determination of the prediction direction of the intra prediction mode indicated by 0, the quantization parameter of the adjacent block on the left or the upper is derived as the prediction quantization parameter. It is also possible to use the derived weighted average value as a predictive quantization parameter.

Assuming that the weighting factor FA of the quantization parameter of the upper neighboring block and the weighting factor FL of the quantization parameter of the left neighboring block are given as threshold values for the prediction direction of the intra prediction mode shown in FIG. 32, TH (= 18) If the intra prediction direction is less than 18, the weighting is set to be large for the quantization parameter of the above adjacent block so that FA> FL, and if it is 18 or more, the left is set so that FA <FL. A large weight is set for the quantization parameter of the adjacent block.

FIG. 33 is a flowchart illustrating the operation of the second prediction quantization parameter deriving unit 304 and is a diagram for explaining the detailed operation of quantization parameter prediction (S709) in the intra prediction direction of FIG.

First, it is determined whether or not the converted intra prediction direction is 0 (S730). The intra prediction direction is 0 when intra prediction based on average value prediction is performed (S7).
30 Yes). In this case, either the left or the upper neighboring block is not preferentially selected, and the weighted coefficients FL and FA of the left and upper neighboring blocks are set to the same value (S731). For example, FA = FL = 2 is set. On the other hand, if the intra prediction direction is not 0 (No in S730), it is determined whether the intra prediction direction is less than 18 (
S732). When the intra prediction direction is less than 18 (Yes in S732), FA> FL
A large weight is set for the quantization parameter of the upper neighboring block so that
S733). For example, FA is set to 3 and FL is set to 1. On the other hand, when the intra prediction direction is 18 or more (No in S732), a large weight is set for the quantization parameter of the left neighboring block so that FA <FL (S734). For example, FA is 1 and FL is 3
Set to.

As described above, the predicted quantization parameter predQP is derived from the determined weighting coefficient and each quantization parameter by the following equation (S735).

Here, the denominator of the above formula is FA + FL, and 2 of the numerator is added for rounding off (FA
+ FL) / 2. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

Here, as a combination of the weighting coefficients FA and FL, (FA, FL) = (3
, 1), (2, 2), (1, 3), but other coefficients may be set. However, when importance is attached to speeding up the calculation, it is desirable to select a variable in which FA + FL is expressed as a power of 2. When the above-described predictive quantization parameter derivation expression is performed by bit shift calculation, it is expressed by

Further, as a combination of the coefficients of the weighting coefficients FA and FL, (FA, FL) = (3, 1)
(FA, FL) = (4, 0), (FA, FL) = (1, 3) into (FA, FL) = (0,
Assuming 4), the former is the upper adjacent block, and the latter is the left adjacent block. This is the same result as in the sixth embodiment. However, in the sixth embodiment, calculation processing is performed only when the average value of QPA and QPL is calculated, and when the average value is not used as the prediction quantization parameter, it is only necessary to select the quantization parameter of the adjacent block on the left or the upper side. Therefore, it is possible to realize simplification and speeding up of processing.

In the sixth embodiment, as a reference determination of the quantization parameter of the adjacent block on the left or upper side,
Although the threshold TH of the intra prediction direction is set to 18, it may not be this value, and can be set implicitly as long as there is no contradiction between the video encoding device 100 and the video decoding device 200. In addition, a threshold value for the intra prediction direction may be set in the header information, and may be changed in units such as a sequence, a picture, and a slice.

Further, the number of thresholds in the prediction direction of the intra prediction mode is increased, and the weighting coefficients of the quantization parameters of the left and upper neighboring blocks are changed and set for each range of the prediction direction region of the intra prediction mode divided by the threshold. You may do it.

FIG. 34 is a diagram illustrating an example in which the prediction direction in the intra prediction mode is divided into three regions by threshold values THLO and THHI. The threshold values THLO and THHI are indicated by thick dotted lines in FIG. 34, and are set to the same weighting coefficient within the range of the region delimited by the threshold values. In the intra prediction mode on the right side of THLO, intra prediction is performed mainly using the pixel signal of the encoded block that is close to the encoded block, and therefore, the correlation with the adjacent block above is high. On the other hand, in the intra prediction mode below THHI, intra prediction is performed mainly using the pixel signal of the encoded block that is close to the left of the encoded block, so the correlation with the left adjacent block is high. high. Therefore, when the intra prediction mode is on the right side of THLO, the weighting is increased for the quantization parameter of the upper neighboring block (FA> FL), and TH
If it is below HI, the weighting is increased for the quantization parameter of the left adjacent block (FA <FL). Also, in the intra prediction mode between THLO and THHI, since it is not biased to one of the encoded blocks that are close to the left and above the encoded block,
Equalize weighting for quantization parameters of left and top neighboring blocks (FA = FL)
. The threshold values THLO and THHI are values in the intra prediction direction.
Is 13 and THHI is 22.

FIG. 35 is a flowchart illustrating the operation of the second predictive quantization parameter deriving unit 304 and explains the detailed operation of the prediction of the quantization parameter (S709) in the intra prediction direction of FIG.

First, it is determined whether or not the converted intra prediction direction is 0 (S740). The intra prediction direction is 0 when intra prediction based on average value prediction is performed (S7).
40 Yes). In this case, either the left or the upper neighboring block is not selected with priority, and the weighted coefficients FL and FA of the left and upper neighboring blocks are set to the same value (S741). For example, FA = FL = 2 is set. On the other hand, when the intra prediction direction is not 0 (No in S740), first, it is determined whether or not the intra prediction direction is less than a threshold value THLO (S742). When the intra prediction direction is less than THLO (S742)
Yes), a larger weight is set for the quantization parameter of the upper adjacent block so that FA> FL (S743). For example, FA is set to 3 and FL is set to 1. When the intra prediction direction is greater than or equal to THLO (No in S742), the intra prediction direction is the threshold value THH.
It is determined whether it is less than I (S744). If the intra prediction direction is not less than THHI, that is, is greater than or equal to THHI (No in S744), a large weight is set for the quantization parameter of the left adjacent block so that FA <FL (S745). For example, FA is 1
, FL is set to 3. When the intra prediction direction is less than THHI (Yes in S744)
), Since the weights for the quantization parameters of the left and upper neighboring blocks are equalized, S7
Proceed to 41.

As described above, the predicted quantization parameter predQP is derived from the determined weighting coefficient and each quantization parameter by the following equation (S746).

Here, the denominator of the above formula is FA + FL, and 2 of the numerator is added for rounding off (FA
+ FL) / 2. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

Here, as a combination of the weighting coefficients FA and FL, (FA, FL) = (3
, 1), (2, 2), (1, 3), but other coefficients may be set. However, when importance is attached to speeding up the calculation, it is desirable to select a variable in which FA + FL is expressed as a power of 2. When the above-described predictive quantization parameter derivation expression is performed by bit shift calculation, it is expressed by

As a combination of the above-described weighting coefficients FA and FL, (FA, FL) = (3, 1
) To (FA, FL) = (4, 0), (FA, FL) = (1, 3) to (FA, FL) = (0
, 4), the former is the upper neighboring block, and the latter is the left neighboring block. In this case, calculation processing is performed to calculate the average value of QPA and QPL in the prediction direction of the intra prediction mode between THLO and THHI, and if the other average values are not used as the prediction quantization parameter, the left or upper proximity This can be realized simply by selecting the quantization parameter of the block. As a result, the processing can be simplified and speeded up.

In the sixth embodiment, the values of the thresholds THLO and THHI are set to 13 and 22 in the intra prediction direction, respectively. However, these values may not be used, and the moving picture coding apparatus 100 and the moving picture decoding apparatus 200 may be different. As long as no contradiction arises, it can be set implicitly.

Furthermore, in the sixth embodiment, the prediction direction of the intra prediction mode is divided into three regions using the threshold values THLO and THHI. However, the number of divisions may be increased by increasing the number of threshold values. However, the weighting coefficients FA and FL coefficients set in each area are FA when importance is placed on speeding up the calculation.
It is desirable to select a variable in which + FL is expressed as a power of 2.

In the sixth embodiment, when the encoded block is close to the upper tree block, it is prohibited to use the quantization parameter of the encoded block in the upper tree block for prediction, but it is close to the left. In the above description, the quantization parameter of the encoded block of the tree block to be used is used for prediction, but the quantization parameter of the encoded block in the left tree block in addition to the above is prohibited from being used for prediction. Even if applicable. In this case, S7
This can be realized simply by replacing the proximity decision of the upper tree block from 01 to S703 from the top to the left and changing the QPA to QPL to the decision branch of S704 to S706.

[Example 7]
The operation of the second predicted quantization parameter derivation units 304 and 205 in the seventh embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 304 to 205, and the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. In the seventh embodiment, as in the sixth embodiment, the quantization parameters of the left and upper encoded blocks close to the encoded block are used for prediction. On the other hand, the difference from the sixth embodiment is that when the encoded block is close to the upper tree block, when the encoded block is close to the left tree block, the quantized blocks in the left tree block are quantized. The use of parameters for prediction is prohibited. This is because the quantization parameter of the coding block is calculated based on the coding processing order of the coding control, so that the coding processing order is separated between the coding blocks between the tree blocks as compared to the inside of the tree block. Even if coding blocks are close to each other between the tree blocks, the quantization parameter of the coding block calculated by the code amount control is not necessarily a close value and may not be suitable as a predictive quantization parameter. is there. Therefore, in the seventh embodiment, similarly to the second embodiment, when the encoding block to be encoded or decoded is close to the left or upper tree block, the encoded block in the left or upper tree block is used. The quantization parameter is not used for prediction, but is used in place of the quantization parameter of the block encoded before or immediately before the encoding block to be encoded in the encoding processing order.

FIG. 36 is an enlarged view of the periphery of the coding block BLK1 in the divided tree block shown in FIG. The direction of the encoded block referred to by BLK1 in FIG. 36 is represented by a thick arrow. The solid thin line in FIG. 36 represents the encoding processing order, and the encoding block uses in principle the quantization parameter of the encoded block that is close within the tree block including the encoding block. Since BLK1 is bordered by the upper tree block, the quantization parameter of the encoded block adjacent to the upper side is not used for prediction, and the block encoded before or immediately before the encoding block to be encoded is used. Is used for prediction together with the quantization parameter of the coded block adjacent to the left. Here, if BLK1 is determined to be intra prediction as the prediction mode and the intra prediction mode is 23, the prediction direction of the intra prediction mode is represented by a thick dotted line in FIG. When the intra prediction mode is 23, since the prediction direction is on the upper side, it is determined to refer to the upper adjacent block. However, since BLK1 is bounded by the upper tree block, it cannot refer to the quantization parameter of the upper neighboring block, and the quantum of the block encoded before or immediately before the encoding block to be encoded that has been replaced. Refer to the conversion parameter. The quantization parameter of the block coded before or immediately before the coding block to be coded is different from the quantization parameter of the upper neighboring block which is actually determined in the prediction direction of the intra prediction mode. Since the quantization parameter is the closest in the processing order, it is possible to suppress deterioration in accuracy of the predicted quantization parameter as much as possible. In addition, by replacing the quantization parameters of the adjacent blocks on the left and top adjacent to the coding block with the quantization parameters of the block coded before or just before the coding block to be coded, the left and top are always changed. The non-zero quantization parameter exists in the adjacent block, and the quantization parameter based on the intra prediction mode depends on whether the left and upper adjacent block quantization parameters exist as in the sixth embodiment. There is no need to switch between prediction and other processing, and no decision branch is required. Therefore, the quantization parameter prediction determination process can be simplified as compared with the sixth embodiment.

FIG. 37 is a flowchart showing the operation of the second predictive quantization parameter deriving unit 304 in the seventh embodiment.

First, the position information of the encoding block to be encoded is acquired (S800). As the position information of the encoded block, the upper left position of the tree block including the encoded block is obtained from the upper left corner of the screen, and the position of the encoded block is obtained from the upper left position of the tree block. It is then determined whether the encoded block is close to the upper tree block (
S801). When the coding block is close to the upper tree block (Yes in S801)
s), that is, when the coding block is located at the upper end of the tree block, since the upper neighboring block is included in the upper tree block, the tree block boundary is exceeded. The quantization parameter prevQP of the coding block coded before or just before the coding block to be coded is set to QPA without using the adjacent block of (S802). On the other hand, if the encoded block is not close to the upper tree block (No in S801), that is, if the upper adjacent block is located in the same tree block as the encoded block, the code is read from the memory 302. The quantization parameter QPA of the already-encoded block adjacent on the quantization block is acquired (S803). At this time, the storage area stored in the encoded information storage memory 113 is accessed based on the reference position information at the upper left of the encoded block, and the quantization parameter of the corresponding surrounding encoded block is the predicted quantization parameter. It is supplied to the memory 302 inside the deriving unit 114, and the quantization parameter of the upper neighboring block is acquired from the memory 302.

Subsequently, it is determined whether or not the encoded block is close to the left tree block (S
804). When the coding block is close to the left tree block (Yes in S804)
), I.e., if the coding block is located at the left end of the tree block, the left neighboring block is included in the left tree block, and thus the tree block boundary is exceeded. Without using adjacent blocks, before the encoding block to be encoded,
Or the quantization parameter prevQP of the coding block coded immediately before is set to QPL (S805). On the other hand, if the coding block is not close to the left tree block (
No in S804), that is, if the left adjacent block is located in the same tree block as the encoded block, the quantization parameter of the already encoded block adjacent to the left of the encoded block from the memory 302 A QPL is acquired (S806).

Next, the intra prediction mode intraPredMo of the coding block is read from the memory 302.
de is acquired and converted into the intra prediction direction intraPredDirec (S807).
. Since the conversion to the intra prediction direction is the same as in the sixth embodiment, the description is omitted. The quantization parameter is predicted based on the converted intra prediction direction (S808). Here, Example 6
Similarly, the threshold TH of the prediction direction of the intra prediction mode shown in FIG. 30 is set to 18 as the value of the intra prediction direction. If the intra prediction direction is less than 18, the quantization parameter of the upper neighboring block is set. If it is 18 or more, the quantization parameter of the left neighboring block is derived as the predicted quantization parameter.

The detailed operation of the prediction of the quantization parameter by the intra prediction direction will be described with reference to FIG. First, it is determined whether or not the converted intra prediction direction is 0 (S810).
The intra prediction direction is 0 when intra prediction based on average value prediction is performed (
Yes in S810). In this case, either the left or upper neighboring block is not selected preferentially, and the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks is set as the predicted quantization parameter (S811). On the other hand, when the intra prediction direction is not 0 (S8
10), it is determined whether or not the intra prediction direction is less than 18 (S812). When the intra prediction direction is less than 18 (Yes in S812), the upper neighboring block is referred to, and the quantization parameter QPA of the upper neighboring block is used as the predicted quantization parameter (S81).
3). On the other hand, when the intra prediction direction is 18 or more (No in S812), the left neighboring block is referred to, and the quantization parameter QPL of the left neighboring block is used as the predicted quantization parameter (S814). The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

As shown in FIG. 38, the prediction of the quantization parameter based on the intra prediction direction of the seventh embodiment (S808) is the prediction of the quantization parameter based on the intra prediction direction of the sixth embodiment (S70).
It is the same as 9). Therefore, in the sixth embodiment, when the weighting coefficient is set to the quantization parameter of the left and upper neighboring blocks according to the prediction direction of the intra prediction mode, FIG.
As shown in FIG. 4, the number of thresholds in the prediction direction of the intra prediction mode is increased, and the weighting coefficient of the quantization parameter of the left and upper neighboring blocks is set for each range of the prediction direction region of the intra prediction mode divided by the threshold. Even in this case, the embodiment 7 can be implemented in the same manner, and the description is omitted here.

[Example 8]
The operation of the second predicted quantization parameter derivation units 304 and 205 in the eighth embodiment will be described. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, the prediction quantization parameter deriving units 304 to 205, and the encoding information storage memory 3
By replacing the code from 02 to 204 and changing the output destination of the predicted quantization parameter from the differential quantization parameter generation unit 111 to the quantization parameter generation unit 203, an equivalent process is realized. The difference from the first embodiment is that, when the encoding block to be encoded or decoded is close to the upper tree block, the encoding in the left tree block is already performed when it is close to the left tree block. The point is to prohibit the use of block quantization parameters for prediction. That is, the quantization parameter of the encoded block that exceeds the tree block boundary is used for prediction because the first encoding block in the tree block is preceded by the encoding block to be encoded. Or, it is limited only when the quantization parameter of the block coded immediately before is used. Further, before the quantization parameter is predicted according to the intra prediction direction, if the left and upper coded neighboring blocks are not positioned beyond the tree block boundary, that is, the left and upper neighboring blocks are quantized. Determination processing when the prediction of the quantization parameter in the intra prediction mode according to the sixth embodiment is not performed by omitting the determination whether the parameter is non-zero and performing the prediction of the quantization parameter in the direct intra prediction mode. (S710 to S714) is reduced.

In the eighth embodiment, when the adjacent block determined in the prediction direction of the intra prediction mode is not located at a position beyond the boundary of the tree block, a prediction quantization parameter is derived from the quantization parameter of the determined adjacent block, and the intra block is calculated. If the adjacent block determined in the prediction direction of the prediction mode is located beyond the boundary of the tree block, the quantization parameter of the determined adjacent block cannot be used, so that quantization of other adjacent blocks is performed. It is determined whether a parameter or a quantization parameter of a block encoded immediately before or immediately before an encoding target encoding block is valid, and is derived as a predicted quantization parameter.

FIG. 39 is a flowchart showing the operation of the first predictive quantization parameter deriving unit 303 in the eighth embodiment. In the flowchart of FIG. 39, S900 to S903 and S907 are S700 to S703 and S708 in FIG. 28 of the sixth embodiment, and S904 to S906 are the seventh embodiment.
37 are the same as S804 to S806 in FIG. 37, and thus the description is omitted, and only the detailed operation of the quantization parameter prediction S908 based on the intra prediction direction will be described.

The detailed operation of the prediction of the quantization parameter based on the intra prediction direction (S908) will be described with reference to FIG. Here, similarly to the sixth embodiment, the threshold TH of the prediction direction in the intra prediction mode shown in FIG. 30 is set to 18 as the value of the intra prediction direction, and the prediction of the quantization parameter is performed based on the threshold TH. And predictive quantization parameters are derived.

First, it is determined whether or not the converted intra prediction direction is 0 (S910). The intra prediction direction is 0 when intra prediction based on average value prediction is performed (S9).
10 Yes). In this case, the process proceeds to method1 (S911). On the other hand, when the intra prediction direction is not 0 (No in S910), it is determined whether or not the intra prediction direction is less than 18 (S912). When the intra prediction direction is less than 18 (Yes in S912), me
The process proceeds to thod2 (S913), and the intra prediction direction is 18 or more (N in S912).
o) Go to method3 (S914).

First, method1 will be described with reference to FIG. In S910, the intra prediction direction is determined to be 0 (Yes in S910) when intra prediction based on average value prediction is performed. In this case, either the left or the upper neighboring block is not preferentially selected, and the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks is used as the predicted quantization parameter. Next, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S920). When the quantization parameters of the left and top neighboring blocks are both positive (Yes in S920), both the left and top neighboring blocks exist. Therefore, since it is possible to obtain the quantization parameter by referring to the neighboring block specified in the direction indicated by the intra prediction mode, the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks Is a predicted quantization parameter (S921).

On the other hand, if the quantization parameter of the left and upper neighboring blocks is not positive (No in S920), that is, the quantization parameter of at least one of the left or upper neighboring blocks is 0, and at least either the left or upper one One of the adjacent blocks does not exist. For this reason, the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks cannot be used as the predicted quantization parameter. Depending on whether or not the quantization parameter is 0, a possible quantization parameter is set as a predicted quantization parameter.

First, it is determined whether or not the quantization parameter of the upper adjacent block is 0 (S922).
. When the quantization parameter of the upper neighboring block is not 0 (No in S922), the quantization parameter of the upper neighboring block is set as a predicted quantization parameter (S923). On the other hand, if it is 0 (Yes in S922), the quantization parameter of the upper neighboring block cannot be referred to as the predicted quantization parameter, and thus the process proceeds to S924. Next, it is determined whether the quantization parameter of the left adjacent block is 0 (S924). When the quantization parameter of the left neighboring block is not 0 (No in S924), the quantization parameter of the left neighboring block is set as the predicted quantization parameter (S925). On the other hand, if it is 0 (Yes in S924), that is, the quantization parameters of the left and upper neighboring blocks are both 0 and neither exists, the quantization of the left and upper neighboring blocks as the prediction quantization parameter. The parameter cannot be referenced. Therefore, the quantization parameter (prevQP) of the block coded before or just before the coding block to be coded is set as the prediction quantization parameter (S926). If the block at the upper left corner of the image is an encoded block, there is no block encoded before and immediately before the encoding block to be encoded, and the adjacent block on the left and upper, so a picture or slice The quantization parameter is used as a predicted quantization parameter.

Next, method 2 will be described with reference to FIG. Since it is determined in S912 that the intra prediction direction is less than 18 (Yes in S912), the prediction parameter is selected as the quantization parameter of the upper coded neighboring block. It is determined whether or not the quantization parameter is 0 (S930). When the quantization parameter of the upper neighboring block is not 0 (No in S930), the quantization parameter of the upper neighboring block is set as the predicted quantization parameter (S931). On the other hand, when it is 0 (Yes in S930), the quantization parameter of the upper adjacent block cannot be referred to as the predicted quantization parameter, and thus the process proceeds to S932. Next, it is determined whether the quantization parameter of the left adjacent block is 0 (S932). When the quantization parameter of the left neighboring block is not 0 (No in S932), the quantization parameter of the left neighboring block is set as the predicted quantization parameter (S933). On the other hand, if it is 0 (Yes in S932), that is, the quantization parameters of the left and upper neighboring blocks are both 0 and neither exists, the quantization of the left and upper neighboring blocks as the prediction quantization parameter. The parameter cannot be referenced. Therefore, the quantization parameter of the block coded before or just before the coding block to be coded (
prevQP) is set as a prediction quantization parameter (S934). If the block at the upper left corner of the image is an encoded block, there is no block encoded before and immediately before the encoding block to be encoded, and the adjacent block on the left and upper, so a picture or slice The quantization parameter is used as a predicted quantization parameter.

Next, method 3 will be described with reference to FIG. Since the intra prediction direction is determined to be 18 or more in S912 (No in S912), the predictive quantization parameter is selected as the quantization parameter of the left encoded neighboring block. It is determined whether or not the quantization parameter is zero (S940). When the quantization parameter of the left neighboring block is not 0 (No in S940), the quantization parameter of the left neighboring block is set as the predicted quantization parameter (S941). On the other hand, if it is 0 (Yes in S940), the quantization parameter of the left adjacent block cannot be referred to as the predicted quantization parameter, and thus the process proceeds to S942. Next, it is determined whether or not the quantization parameter of the upper neighboring block is 0 (S942). When the quantization parameter of the upper neighboring block is not 0 (No in S942), the quantization parameter of the upper neighboring block is set as the predicted quantization parameter (S943). On the other hand, if it is 0 (Yes in S942), that is, the quantization parameters of the left and upper neighboring blocks are both 0 and neither exists, the quantization of the left and upper neighboring blocks as the prediction quantization parameter. The parameter cannot be referenced. Therefore, the quantization parameter of the block coded before or just before the coding block to be coded (
prevQP) is set as a prediction quantization parameter (S944). If the block at the upper left corner of the image is an encoded block, there is no block encoded before and immediately before the encoding block to be encoded, and the adjacent block on the left and upper, so a picture or slice The quantization parameter is used as a predicted quantization parameter.

In each method branched in the intra prediction direction as described above, when the quantization parameter of the adjacent block determined in the intra prediction direction is 0, the encoding block to be encoded is performed without performing the subsequent determination. It is also possible to simplify the determination process using the quantization parameter (prevQP) of the block encoded before or immediately before as the prediction quantization parameter. In method 1, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S920). If the quantization parameters of the left and upper neighboring blocks are not positive (No in S920), S922 is determined. The quantization parameter (prevQP) of the block encoded before or immediately before the encoding block to be encoded without performing the subsequent determination process.
) As a predicted quantization parameter (S926). In method 2, it is determined whether or not the quantization parameter of the upper adjacent block is 0 (S 930). If it is 0 (Yes in S 930), the encoding process is performed without performing the determination process after S 932. The quantization parameter (prevQP) of the block encoded before or immediately before the block is set as the predicted quantization parameter (S934). In method 3, the quantization parameter of the left neighboring block is 0.
If it is 0 (Yes in S940), the quantum of the block encoded before or immediately before the encoding block to be encoded without performing the determination processing after S942 is determined. The quantization parameter (prevQP) is set as a predicted quantization parameter (S944).
The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

The prediction of the quantization parameter based on the intra prediction direction of the eighth embodiment (S908) is performed according to the prediction direction of the intra prediction mode, similarly to the prediction of the quantization parameter based on the intra prediction direction of the sixth embodiment (S709). When the weighting coefficient is set to the quantization parameter of the adjacent block above, the number of thresholds in the prediction direction of the intra prediction mode is further increased as shown in FIG. This can also be implemented when setting the weighting coefficient of the quantization parameter of the left and upper neighboring blocks for each area range. Here, as shown in FIG. 34, the number of thresholds in the prediction direction of the intra prediction mode is increased, and the quantization parameter of the adjacent block on the left and the upper is separated for each range of the prediction direction region of the intra prediction mode divided by the threshold. As an example of setting a weighting coefficient, a case where the present invention is applied to quantization parameter prediction based on the intra prediction direction (S908) will be described. Other cases described in the sixth embodiment are also possible in the same manner as in the case where the eighth embodiment described below is applied, and thus the description thereof is omitted.

FIG. 44 shows an increase in the number of thresholds in the prediction direction of the intra prediction mode as shown in FIG. 5 is a flowchart showing detailed operation of quantization parameter prediction based on the intra prediction direction (S908) when setting the weighting coefficient. Example 6
Similarly to the above, the prediction direction of the intra prediction mode shown in FIG.
Divide into three regions, THLO is set to 13 and THHI is set to 22, respectively, and the prediction of the quantization parameter determined for each region in which the intra prediction direction is divided by a threshold is switched. Is to be derived.

First, it is determined whether or not the converted intra prediction direction is 0 (S950). The intra prediction direction is 0 when intra prediction based on average value prediction is performed (S9).
50 Yes). In this case, the process proceeds to method1 (S951). On the other hand, when the intra prediction direction is not 0 (No in S950), it is determined whether the intra prediction direction is less than THLO (S952). When the intra prediction direction is less than THLO (Yes in S952)
) And method2 (S953). When the intra prediction direction is greater than or equal to THLO (
(No in S952), it is determined whether the intra prediction direction is less than the threshold THHI (S954).
). When the intra prediction direction is not less than THHI, that is, is equal to or higher than THHI (No in S954), the process proceeds to method3 (S955). If the intra prediction direction is less than THHI (Yes in S954), the weighting for the quantization parameter of the left and upper neighboring blocks is equalized, so the process proceeds to S951.

Next, detailed operation of method2 (S953) will be described. FIG. 45 shows method
It is a flowchart explaining the detailed operation | movement of od2 (S953). When determining the weighted average, both the left and top neighboring blocks must be present. Therefore, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S960). When the quantization parameters of the left and upper neighboring blocks are both positive (Yes in S960), since both the left and upper neighboring blocks exist, the upper neighboring block is quantized so that FA>FL> 0. A large weight is set for the parameter (S961). For example, FA is 3
, FL is set to 1. On the other hand, if the quantization parameter of the left and upper neighboring blocks is not positive (No in S960), that is, the quantization parameter of at least one of the left or upper neighboring blocks is 0, and at least either the left or upper one One of the adjacent blocks does not exist. Therefore, the weighted average cannot be obtained from the quantization parameters of the left and upper neighboring blocks, so the presence or absence of the left or upper neighboring block, that is, the quantization parameter of the left or upper neighboring block is 0. Depending on whether or not, a possible quantization parameter is set as a predicted quantization parameter. First, it is determined whether or not the quantization parameter of the upper neighboring block is 0 (S962).

If the quantization parameter of the upper neighboring block is not 0 (No in S962), the weighting coefficient FL of the quantization parameter of the left neighboring block is set to be the predicted quantization parameter. It is set to 0 (S963). For example, FA is 4, FL
Is set to 0. On the other hand, if it is 0 (Yes in S962), the quantization parameter of the above adjacent block cannot be referred to as the predicted quantization parameter, and thus the process proceeds to S964. Next, it is determined whether or not the quantization parameter of the left adjacent block is 0 (S964). If the quantization parameter of the left neighboring block is not 0 (No in S964), the weighted coefficient FA of the quantization parameter of the upper neighboring block is set to be the predicted quantization parameter. It is set to 0 (S965). For example, FA is 0, FL is 4
Set to. On the other hand, when it is 0 (Yes in S964), that is, the quantization parameters of the left and upper neighboring blocks are both 0, and both of them do not exist. The parameter cannot be referenced. Therefore, the quantization parameter (pr of the block encoded before or immediately before the encoding block to be encoded)
evQP) is set as a prediction quantization parameter (S967). When the weighted coefficient of the quantization parameter of the left and upper neighboring blocks is set, the predicted quantization parameter predQP is derived from the determined weighting coefficient and each quantization parameter by the following equation (S966).

Here, the denominator of the above formula is FA + FL, and 2 of the numerator is added for rounding off (FA
+ FL) / 2. The predicted quantization parameter derived in this way is supplied to the differential quantization parameter generation unit 111.

For method1 (S951) and method3 (S955), method is also used.
This is realized by a determination process similar to d2. However, the setting of the weighted coefficients of the quantization parameters of the left and upper neighboring blocks in S961 is FA = FL ≠ 0, met in method1.
In hod3, it is necessary to set the weighting so that 0 <FA <FL.

Here, coefficients other than the combination of the weighting coefficients FA and FL may be set.
However, when importance is attached to speeding up the calculation, it is desirable to select a variable in which FA + FL is expressed as a power of 2. When the above-described predictive quantization parameter derivation expression is performed by bit shift calculation, it is expressed by

As a combination of the above-described weighting coefficients FA and FL, (FA, FL) = (3, 1
) To (FA, FL) = (4, 0), (FA, FL) = (1, 3) to (FA, FL) = (0
, 4), the former is the upper neighboring block, and the latter is the left neighboring block. In this case, calculation processing is performed to calculate the average value of QPA and QPL in the prediction direction of the intra prediction mode between THLO and THHI, and if the other average values are not used as the prediction quantization parameter, the left or upper proximity This can be realized simply by selecting the quantization parameter of the block. As a result, the processing can be simplified and speeded up.

In the eighth embodiment, the values of the thresholds THLO and THHI are set to 13 and 22 in the intra prediction direction, respectively, but these values may not be used. As long as no contradiction arises, it can be set implicitly.

Furthermore, in the eighth embodiment, the prediction direction of the intra prediction mode is divided into three regions with the thresholds THLO and THHI. However, the number of divisions may be increased by increasing the number of thresholds. However, the weighting coefficients FA and FL coefficients set in each area are FA when importance is placed on speeding up the calculation.
It is desirable to select a variable in which + FL is expressed as a power of 2.

[Example 9]
In the sixth and eighth embodiments, there is a case where the adjacent block determined based on the prediction direction of the intra prediction mode of the coding block to be coded is located beyond the boundary of the tree block. In the sixth embodiment, in order to enable prediction of the quantization parameter in the intra prediction mode, it is determined in advance whether or not the left and upper neighboring blocks are both in positions exceeding the boundary of the tree block, and both are exceeded. Predict quantization parameter in intra prediction mode only when not in position, otherwise predict prediction of quantization parameter in intra prediction mode (if at least one neighboring block is beyond the boundary of tree block) Was not implemented. In the eighth embodiment, when the adjacent block determined in the prediction direction of the intra prediction mode is located beyond the boundary of the tree block, the quantization parameter of the determined adjacent block cannot be used. Next, it is determined whether the quantization parameter of the neighboring block or the quantization parameter of the block coded before or just before the coding block to be coded is valid, and is derived as the predicted quantization parameter. As described above, when the adjacent block determined in the prediction direction of the intra prediction mode of the coding block to be encoded is located beyond the boundary of the tree block, the prediction direction of the intra prediction mode and the proximity of the reference destination Since the position of the block does not match, there is a possibility that the accuracy of the determination by the intra prediction mode is lowered. Even if the originally encoded block is intra-prediction, there may be a case where a neighboring block that has already been encoded is located at a position beyond the boundary of the tree block and cannot be determined in the intra-prediction mode. Thus, in the ninth embodiment, when the second prediction quantization parameter derivation unit does not match the determination in the intra prediction mode with the position of the adjacent block to be referenced, the determination result in the intra prediction mode is not adopted, and the first prediction quantum By performing redetermination in the quantization parameter deriving unit, the prediction accuracy of the quantization parameter is improved. Although the encoding process will be described here, in the case of the decoding process, the encoding is decoding, and the prediction quantization parameter derivation unit is set to 114.
205, the coding information storage memory is replaced with codes 113 to 204, and the output destination of the predicted quantization parameter is changed from the difference quantization parameter generation unit 111 to the quantization parameter generation unit 20
By setting the number to 3, equivalent processing is realized.

FIG. 46 shows a detailed configuration of the predictive quantization parameter deriving units 114 and 205 to which the determination of the ninth embodiment is applied. The predicted quantization parameter derivation units 114 and 205 include a switch 401, a memory 402, a first predicted quantization parameter derivation unit 403, a second predicted quantization parameter derivation unit 404, and a recalculation determination unit 405. 46 have the same functions as those of the first predictive quantization parameter deriving unit 303 from the switch 301 of FIG. I will omit the explanation. Figure 46
Since the second predictive quantization parameter deriving unit 404 is changed to a function different from that of the second predictive quantization parameter deriving unit 304 of FIG. 10 described above, the difference will be described.

47 is a flowchart for explaining the operation of the second predictive quantization parameter deriving unit 404 in which the function of the ninth embodiment is applied to the second predictive quantization parameter deriving unit 304 described in the sixth embodiment. . The processing from S1000 to S1009 is the same as that of the second predictive quantization parameter deriving unit 304 described above, and a description thereof will be omitted. The difference from the second predicted quantization parameter deriving unit 304 described above is whether or not the quantization parameters of the left and upper neighboring blocks are both positive (S1007), and the quantum of the left and upper neighboring blocks is determined. When at least one of the quantization parameters is not positive (No in S1007), the prediction quantization parameter (predQP) is set to 0 without performing the prediction of the quantization parameter in the intra prediction mode (
S1010). In the second predictive quantization parameter deriving unit 304, even if the encoded block is intra-predicted, neighboring encoded neighboring blocks are at positions beyond the boundary of the tree block, and the quantization parameter of the adjacent block is set. When it is set to 0 and the quantization parameter cannot be predicted in the intra prediction mode, the determination is made based on the presence or absence of the adjacent block regardless of the intra prediction mode. Second predictive quantization parameter deriving unit 40
4, only the prediction of the quantization parameter in the intra prediction mode is performed, and if the determination cannot be made by the prediction of the quantization parameter in the intra prediction mode, the recalculation determination unit 405 described later derives the first prediction quantization parameter. The prediction quantization parameter is set to 0 in order to cause the unit 403 to perform redetermination. The derived prediction quantization parameter is supplied to the recalculation determination unit 405.

FIG. 48 shows the second predicted quantum obtained by applying the function of the ninth embodiment to the prediction of the quantization parameter based on the intra prediction direction of the second predicted quantization parameter deriving unit 304 described in the eighth embodiment (S908). 10 is a flowchart for explaining an operation of predicting a quantization parameter based on an intra prediction direction of the quantization parameter deriving unit 404. As in the eighth embodiment, the threshold TH of the prediction direction in the intra prediction mode shown in FIG. 30 is set to 18 as the value of the intra prediction direction, and the prediction of the quantization parameter is switched according to the threshold TH. Let us derive the predictive quantization parameter. First, it is determined whether or not the converted intra prediction direction is 0 (S1020). The intra prediction direction is 0 when intra prediction based on average value prediction is performed (Yes in S1020). In this case, method
1 (S1021). On the other hand, when the intra prediction direction is not 0 (No in S1020)
Then, it is determined whether or not the intra prediction direction is less than 18 (S1022). Intra prediction direction is 1
If it is less than 8 (Yes in S1022), the process proceeds to method2 (S1023). If the intra prediction direction is 18 or more (No in S1022), method3 (S1024)
)

First, method1 will be described with reference to FIG. In S1020, the intra prediction direction is determined to be 0 (Yes in S1020) when intra prediction based on average value prediction is performed. In this case, either the left or the upper neighboring block is not preferentially selected, and the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks is used as the predicted quantization parameter. Next, it is determined whether the quantization parameters of the left and upper neighboring blocks are both positive (S1030). When the quantization parameters of the left and top neighboring blocks are both positive (Yes in S1030), both the left and top neighboring blocks exist. Therefore, referring to the neighboring block specified in the direction shown in the intra prediction mode,
Since the quantization parameter can be obtained, the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks is set as the predicted quantization parameter (S1031). On the other hand, when the quantization parameters of the left and upper neighboring blocks are not positive (N in S1030)
o), that is, the quantization parameter of at least one of the adjacent blocks on the left or above is 0
Therefore, at least one of the adjacent blocks on the left or upper side does not exist.
For this reason, the average value of the quantization parameters QPL and QPA of the left and upper neighboring blocks cannot be used as the prediction quantization parameter, so the prediction quantization parameter is set to 0 (S10
32).

Next, method 2 will be described with reference to FIG. Since it is determined in S1022 that the intra prediction direction is less than 18 (Yes in S1022), the predicted quantization parameter is selected as the quantization parameter of the upper coded neighboring block. It is determined whether or not the quantization parameter is 0 (S1040). When the quantization parameter of the upper neighboring block is not 0 (No in S1040), the quantization parameter of the upper neighboring block is set as the predicted quantization parameter (S1041). On the other hand, if it is 0 (
Since the quantization parameter of the above adjacent block cannot be referred to as the predicted quantization parameter, the predicted quantization parameter is set to 0 (S1042).

Next, method 3 will be described with reference to FIG. Since the intra prediction direction is determined to be 18 or more in S1022 (No in S1022), the quantization parameter of the left encoded neighboring block is selected as the prediction quantization parameter. It is determined whether or not the quantization parameter is 0 (S1050). When the quantization parameter of the left neighboring block is not 0 (No in S1050), the quantization parameter of the left neighboring block is set as the predicted quantization parameter (S1051). On the other hand, if it is 0 (
Since the quantization parameter of the left neighboring block cannot be referred to as the prediction quantization parameter, the prediction quantization parameter is set to 0 (S1052).

As described above, in each method branched in the intra prediction direction, when the quantization parameter of the adjacent block determined in the intra prediction direction is 0, the adjacent block is in a position beyond the boundary of the tree block. Or it is a case where it is outside a screen and a prediction quantization parameter cannot be derived | led-out by the quantization parameter of the adjacent block determined by the intra prediction direction. Therefore, the prediction quantization parameter is set to 0 in order to indicate that the determination cannot be made by the prediction of the quantization parameter in the intra prediction mode. The derived prediction quantization parameter is supplied to the recalculation determination unit 405.

Thus, the predicted quantization parameter (predQP) derived by the second predicted quantization parameter deriving unit 404 is supplied to the recalculation determining unit 405.

The detailed operation of the recalculation determination unit 405 will be described with reference to FIG. The recalculation determination unit 405 is supplied with the predicted quantization parameter from the second predicted quantization parameter derivation unit 404 (S
1100). The recalculation determination unit 405 determines whether or not the supplied prediction quantization parameter is 0 (S1101). If the predicted quantization parameter is 0, it is considered that the second predicted quantization parameter deriving unit 404 has not obtained an appropriate determination result, and the first predicted quantization parameter deriving unit 403 derives the predicted quantization parameter again. (S1102). When the predicted quantization parameter is not 0, the predicted quantization parameter derived by the second predicted quantization parameter deriving unit 404 is supplied to the differential quantization parameter generating unit 111.

As described above, when the second prediction quantization parameter deriving unit does not match the determination in the intra prediction mode with the position of the adjacent block to be referenced, the determination result in the intra prediction mode is not adopted, and the first prediction quantization is performed. Re-determination can be performed by the parameter deriving unit.

In the ninth embodiment, when the second prediction quantization parameter deriving unit 404 does not match the determination in the intra prediction mode with the position of the adjacent block to be referenced, the determination result in the intra prediction mode is not adopted, and the prediction quantization parameter is set to 0. In other words, the quantization parameter prediction determination by the intra prediction mode of the second prediction quantization parameter deriving unit 304 of the sixth and eighth embodiments is not changed. Therefore, in the sixth embodiment, when the weighting coefficient is set to the quantization parameter of the left and upper neighboring blocks according to the prediction direction of the intra prediction mode, the prediction direction of the intra prediction mode is further illustrated as shown in FIG. In the case of setting the weighting coefficient of the quantization parameter of the left and upper neighboring blocks for each range of the prediction direction area of the intra prediction mode divided by the threshold, Since it is possible, the explanation is omitted here.

According to the moving image encoding apparatus of the embodiment, the quantization parameter encoded for each block to be encoded is predicted using the quantization parameter of the surrounding encoded block. Deriving the optimal prediction quantization parameter, encoding the difference between the quantization parameter and the prediction quantization parameter, and reducing the coding amount of the quantization parameter without changing the image quality. Efficiency can be improved.

Further, since the encoding side and the decoding side can be implemented as a common function for predicting the quantization parameter, the circuit scale can be reduced. This is because the encoded neighboring block is a locally decoded block for prediction of the next encoded block on the encoding side and is the same as the decoded block, so there is no contradiction between the encoding side and the decoding side. As described above, the determination of the quantization parameter prediction is realized.

In the above description, the prediction of the quantization parameter is performed in units of coding blocks. However, when the number of blocks in the tree block is increased and a large number of coding blocks having a small block size are generated, the coding block in the code amount control is used. In some cases, the assigned code amount per hit becomes too small, and the quantization parameter is not appropriately calculated. In addition, the moving picture coding apparatus 100 and the moving picture decoding apparatus 200 that store coding information such as a quantization parameter at the time of coding and decoding.
This also increases the amount of memory of the encoded information storage memories 113 and 204. Therefore, a block called a quantization group may be newly set as a unit for encoding and transmitting the quantization parameter, and the quantization parameter may be predicted for this block unit.

The quantization group is a block determined according to the size of the tree block, and the size is represented by a value obtained by multiplying the length of the block side of the tree block by 1 / 2n times (n is an integer of 0 or more). That is, a value obtained by shifting the length of the side of the block of the tree block to the right by n bits becomes the length of the side of the quantization group. Since this value determines the block size in the same manner as the tree block structure, it has a high affinity with the tree block. Further, since the tree block is divided into equal sizes, management and readout of the quantization parameters stored in the encoded information storage memories 113 and 204 can be simplified.

FIG. 53 shows an example in which the inside of a tree block is divided by a tree block structure. The block size of the tree block is 64 × 64, the inside of the tree block is hierarchically divided into four, 32 × 32 blocks (dotted rectangle in FIG. 53) in the first division, and 16 × 16 blocks in the second division ( The hatched rectangle in FIG. 53) 8 × 8 blocks in the third division (open rectangle in FIG. 53)
Divided into a plurality of encoded blocks. Here, assuming that the quantization group is a 16 × 16 rectangular block, the quantization group is represented by a thick dotted line in FIG. 53, and the quantization parameter is predicted for each quantization group.

When the encoding block to be encoded is larger than the block size of the quantization group (3
2 × 32 blocks), for example, the inside of a coding block represented by a dotted rectangle in FIG. 53 is divided into four by a quantization group. Although it is divided into four by the quantization group, there is only one quantization parameter for this coding block. Therefore, when the size of the coding block is larger than the quantization group, after the prediction of the quantization parameter of the coding block Are encoded and transmitted, and the same quantization parameter is stored in the memory areas of the encoded information storage memories 113 and 204 corresponding to the respective four quantization groups. Although the quantization parameter overlaps in the memory, it becomes easy to access the quantization parameter of the surrounding encoded block in the prediction of the quantization parameter.

When the encoding block to be encoded is the same as the block size of the quantization group (16 × 1
6 blocks), which is the same as the case of the prediction of the quantization parameter in units of the coding block described above.

When the encoding block to be encoded is smaller than the quantization group block size (8 ×
8 blocks), for example, coding blocks represented by white rectangles in FIG. 53. Since four coding blocks are accommodated in the quantization group, the coding blocks in the quantization group are individually Instead of providing a quantization parameter, one quantization parameter is provided in the quantization group, and individual coding blocks are encoded in the quantization parameter.
In addition, as a quantization parameter of the quantization group, there is a method of selecting one of the quantization parameters of the four coding blocks in the quantization group as a representative value and calculating an average value, but there is no particular limitation here.

FIG. 54 shows an example of quantization parameter prediction when the encoded block is smaller than the block size of the quantization group. In FIG. 54, the hatched rectangle indicates the encoding block to be encoded, the gray rectangle indicates the encoded block used by the quantization group including the encoding block for prediction of the quantization parameter, and the thin solid line indicates the encoding processing order. Represents. The prediction of the quantization parameter is performed with reference to the pixel position at the upper left corner of the quantization group to be processed. When the quantization parameter of the encoded block close to the top is used for prediction, the hatched rectangle in FIG. 54 is 1 for the pixel in the upper left corner of the quantization group including the encoding block to be encoded. The position of the encoded block including the pixel adjacent to the pixel is calculated, and the quantization parameter stored at the address corresponding to the position is called from the encoded information storage memories 113 and 204. Similarly, when the quantization parameter of an encoded block close to the left is used for prediction, the pixel at the upper left corner of the quantization group including the encoding block to be encoded is close to the left by one pixel. The position of the encoded block including the pixel to be calculated is calculated, and the quantization parameter recorded at the address corresponding to the position is called from the encoded information storage memories 113 and 204.
When the encoded block including the pixel adjacent to the left and the upper left pixel of the quantization group including the encoding block to be encoded exceeds the boundary of the tree block, encoding of the encoding target is performed. Since the quantization parameter of the encoded block encoded before or immediately before the block is used, the memory address stored when the quantization parameter is stored in the encoded information storage memories 113 and 204 by encoding is temporarily stored. The quantization parameter stored in advance and stored at the address corresponding to the position before or immediately before the encoding block to be encoded is called from the encoding information storage memories 113 and 204. In this way, it is possible to predict the quantization parameter of the encoding block to be encoded.

As described above, the prediction of the quantization parameter for each quantization group can be performed in the same manner as the prediction of the quantization parameter for each coding block.

Note that the block size of the quantization group may be described directly in the header information of the bitstream, or a bit shift indicating whether to make the block size 1 / 2n times (n is an integer of 0 or more) the tree block size. An amount may be described. For example, a flag cu_qp_delta_enabl that specifies whether or not to perform quantization parameter prediction for each picture in the header information of a picture and describe and transmit the differential quantization parameter in the bitstream
Parameter d that determines the size of the quantization group only when e_flag is defined and flag cu_qp_delta_enable_flag is enabled (set to "1")
If_cu_qp_delta_depth is described in the bitstream. The size of the quantization group is changed from exponent n to dif when the size of the tree block is 2n.
This is expressed as a power of 2 with an index obtained by subtracting f_cu_qp_delta_depth. In addition, the quantization group size may be determined implicitly by encoding and decoding without being described in the bitstream.

The moving image encoded stream output from the moving image encoding apparatus of the embodiment described above has a specific data format so that it can be decoded according to the encoding method used in the embodiment. Therefore, the moving picture decoding apparatus corresponding to the moving picture encoding apparatus can decode the encoded stream of this specific data format.

When a wired or wireless network is used to exchange an encoded stream between a moving image encoding device and a moving image decoding device, the encoded stream is converted into a data format suitable for the transmission form of the communication path. It may be transmitted. In that case, a video transmission apparatus that converts the encoded stream output from the video encoding apparatus into encoded data in a data format suitable for the transmission form of the communication channel and transmits the encoded data to the network, and receives the encoded data from the network Then, a moving image receiving apparatus that restores the encoded stream and supplies the encoded stream to the moving image decoding apparatus is provided.

The moving image transmitting apparatus is a memory that buffers the encoded stream output from the moving image encoding apparatus, a packet processing unit that packetizes the encoded stream, and transmission that transmits the packetized encoded data via the network. Part. The moving image receiving apparatus generates a coded stream by packetizing the received data, a receiving unit that receives the packetized coded data via a network, a memory that buffers the received coded data, and packet processing. And a packet processing unit provided to the video decoding device.

The above processing relating to encoding and decoding can be realized as a transmission, storage, and reception device using hardware, and is stored in a ROM (Read Only Memory), a flash memory, or the like. It can also be realized by firmware or software such as a computer. The firmware program and software program can be recorded on a computer-readable recording medium, provided from a server through a wired or wireless network, or provided as a data broadcast of terrestrial or satellite digital broadcasting Is also possible.

The present invention has been described based on the embodiments. The embodiments are exemplifications, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are within the scope of the present invention. .

DESCRIPTION OF SYMBOLS 100 moving image encoder, 101 image memory, 102 residual signal production | generation part, 1
03 orthogonal transform / quantization unit, 104 second encoded bit string generation unit, 105 inverse quantization / inverse orthogonal transform unit, 106 decoded image signal superimposing unit, 107 decoded image memory, 10
8 prediction image generation unit, 109 activity calculation unit, 110 quantization parameter calculation unit, 111 differential quantization parameter generation unit, 112 first encoded bit string generation unit, 113 encoded information storage memory, 114 prediction quantization parameter derivation unit , 115
An encoded bit string multiplexing unit, 200 moving image decoding apparatus, 201 bit string separating unit,
202 first encoded bit sequence decoding unit, 203 quantization parameter generation unit, 204 encoded information storage memory, 205 predicted quantization parameter derivation unit, 206 second encoded bit sequence decoding unit, 207 inverse quantization / inverse orthogonal transform unit, 208 decoded image signal superimposing unit,
209 predicted image generation unit, 210 decoded image memory, 301 switch, 302
Memory 303 first prediction quantization parameter derivation unit 304 second prediction quantization parameter derivation unit 401 switch 402 memory 403 first prediction quantization parameter derivation unit 404 second prediction quantization parameter derivation Part, 405 recalculation judgment part.

Claims (3)

A moving picture decoding apparatus for decoding a bit stream in which a moving picture is encoded by dividing a first block obtained by dividing each picture of a moving picture into a predetermined size into one or a plurality of second blocks. There,
A decoding unit for extracting block size information for decoding the bitstream and deriving a predictive quantization parameter and a differential quantization parameter of the second block;
Using the quantization parameters of the third block adjacent to the left of the second block and the fourth block adjacent to the second block, the predicted quantization parameter of the second block is derived. A predictive quantization parameter derivation unit;
A quantization parameter generation unit that generates a quantization parameter of the second block by adding the differential quantization parameter of the second block and the predicted quantization parameter;
The predictive quantization parameter derivation unit, when the third block is at a position not exceeding the boundary of the first block, sets the quantization parameter of the third block as the first quantization parameter; If there is a position beyond the boundary of the first block, the quantization parameter of the fifth block decoded immediately before the second block in the decoding order is set as the first quantization parameter, and the fourth block When the block is in a position not exceeding the boundary of the first block, the quantization parameter of the fourth block is set as the second quantization parameter, and the block is located beyond the boundary of the first block. In this case, the quantization parameter of the fifth block is set as a second quantization parameter, and the first quantization parameter and the second quantization parameter are set for each predetermined size based on the size information. Using a quantization parameter to derive a predicted quantization parameter of the second block,
The moving picture decoding apparatus, wherein the decoding unit decodes the differential quantization parameter of the second block so that the code length is shorter as the absolute value is smaller. A moving picture decoding method for decoding a bit stream in which a moving picture is encoded by dividing a first block obtained by dividing each picture of a moving picture by a predetermined size into one or a plurality of second blocks. There,
A decoding step of extracting block size information for decoding the bitstream and deriving a predictive quantization parameter and a differential quantization parameter of the second block;
Using the quantization parameters of the third block adjacent to the left of the second block and the fourth block adjacent to the second block, the predicted quantization parameter of the second block is derived. A predictive quantization parameter derivation step;
A quantization parameter generating step of generating a quantization parameter of the second block by adding the differential quantization parameter of the second block and the predicted quantization parameter;
The predictive quantization parameter deriving step determines the quantization parameter of the third block as the first parameter when the third block is in a position not exceeding the boundary of the first block.
If the position exceeds the boundary of the first block, the quantization parameter of the fifth block decoded immediately before the second block in decoding order is the first quantization parameter. And when the fourth block is in a position not exceeding the boundary of the first block, the quantization parameter of the fourth block is set as a second quantization parameter,
When the position is beyond the boundary of the first block, the quantization parameter of the fifth block is set as the second quantization parameter, and the first quantum is set for each predetermined size based on the size information. Deriving the predicted quantization parameter of the second block using the quantization parameter and the second quantization parameter;
The moving picture decoding method characterized in that the decoding step decodes the differential quantization parameter of the second block so that the code length is shorter as the absolute value is smaller. A moving picture decoding program for decoding a bit stream in which a moving picture is encoded by dividing a first block obtained by dividing each picture of a moving picture into a predetermined size into one or a plurality of second blocks. There,
A decoding step of extracting block size information for decoding the bitstream and deriving a predictive quantization parameter and a differential quantization parameter of the second block;
Using the quantization parameters of the third block adjacent to the left of the second block and the fourth block adjacent to the second block, the predicted quantization parameter of the second block is derived. A predictive quantization parameter derivation step;
Causing the computer to execute a quantization parameter generation step of generating a quantization parameter of the second block by adding the differential quantization parameter of the second block and the predicted quantization parameter;
The predictive quantization parameter deriving step determines the quantization parameter of the third block as the first parameter when the third block is in a position not exceeding the boundary of the first block.
If the position exceeds the boundary of the first block, the quantization parameter of the fifth block decoded immediately before the second block in decoding order is the first quantization parameter. And when the fourth block is in a position not exceeding the boundary of the first block, the quantization parameter of the fourth block is set as a second quantization parameter,
When the position is beyond the boundary of the first block, the quantization parameter of the fifth block is set as the second quantization parameter, and the first quantum is set for each predetermined size based on the size information. Deriving the predicted quantization parameter of the second block using the quantization parameter and the second quantization parameter;
The moving picture decoding program characterized in that the decoding step decodes the differential quantization parameter of the second block so that the code length is shorter as the absolute value is smaller.

Images