Technical field
[0001]
The present disclosure relates to an image processing apparatus and method, and more particularly, to an image processing apparatus and method which make it possible to suppress the reduction of the coding efficiency.
BACKGROUND
[0002]
Conventionally, luminance, TU (Transform Unit) per unit, the primary conversion PThor (also referred to as primary horizontal conversion) in the horizontal direction and in each vertical direction of the primary conversion PTver (also referred to as primary vertical conversion) adaptively multiple different from the orthogonal transform, adaptive primary transform to select the primary translation (AMT: Adaptive Multiple Core transforms) has been disclosed (e.g., see non-Patent Document 1).
CITATION
Non-patent literature
[0003]
非特許文献1 : Jianle Chen, Elena Alshina, Gary J. Sullivan, Jens-Rainer, Jill Boyce, "Algorithm Description of Joint Exploration Test Model 4", JVET-D1001_v3, Joint Video Exploration Team (JVET) of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC 29/WG 11 4th Meeting: Chengdu, CN, 15-21 October 2016
Summary of the Invention
Problems that the Invention is to Solve
[0004]
　However, in Non-Patent Document 1, although only the adaptive primary transform to the luminance (Y) can be applied, not applied to the color difference (Cb, Cr). Therefore, coding efficiency of the primary converter for the color difference, there is a risk of reduced compared to luminance.
[0005]
　The present disclosure has been made in view of such circumstances, it is desirable to make it possible to suppress the reduction of the coding efficiency.
Means for Solving the Problems
[0006]
　The image processing apparatus of the first embodiment of the present technology, by using the information about the inverse orthogonal transform of the color difference which is derived based on the information for the inverse orthogonal transform of the luminance, the inverse orthogonal transform unit for performing the inverse orthogonal transformation of the color difference an image processing apparatus comprising.
[0007]
　Information to include an adaptive primary conversion flag indicating whether to apply the adaptive inverse primary transform is used as adaptively selecting and inverse primary transform one from among a plurality of different inverse orthogonal transform related to the inverse orthogonal transform be able to.
[0008]
　The value of the adaptive primary conversion flag of the color difference can be to be set to the value of the adaptive primary conversion flag brightness.
[0009]
　Information relating to the inverse orthogonal transform, it is possible to include a primary conversion identifier indicating whether to apply any inverse primary transform to the inverse primary transform in the vertical direction and the horizontal direction.
[0010]
　The value of the primary conversion identifier of the color difference, when the adaptive primary conversion flag of the color difference is true, is set to the value of the primary conversion identifier of luminance, if the adaptive primary conversion flag of the color difference is false, set to a predetermined value it can be made to be.
[0011]
　The inverse orthogonal transform unit, as the inverse orthogonal transform, it is possible to perform the inverse primary transform.
[0012]
　The inverse orthogonal transform unit, as the inverse orthogonal transform, it is possible to perform the inverse primary horizontal transform is the inverse primary transform in the horizontal direction, the inverse primary vertical conversion which is the inverse primary transform in the vertical direction.
[0013]
　The value of the adaptive primary conversion flag of the color difference, when the prediction type of the coding block transform block to be processed belongs is an inter prediction, it is possible to be set to the value of the adaptive primary conversion flag luminance .
[0014]
　The value of the adaptive primary conversion flag of color difference, or the prediction type of the coding block transform block to be processed belongs is an inter prediction, or, if the prediction mode is the intra prediction of an intra block copy, the adaptive luminance it is possible to be set to the value of the primary conversion flag.
[0015]
　If the value of the adaptive primary conversion flag of color difference, or the prediction type of the coding block transform block to be processed belongs is an inter prediction or an intra prediction is prediction mode in the luminance and color difference matches, luminance it can be made to be the set of the values ​​of the adaptive primary conversion flag.
[0016]
　The value of the adaptive primary conversion flag of the color difference, when the adaptive primary conversion flag brightness said adaptive primary conversion flag chrominance adaptive primary conversion information estimation flag indicating estimated based on the color difference is true, the adaptive luminance it is possible to be set to the value of the primary conversion flag.
[0017]
　The value of the adaptive primary conversion flag of the color difference, when the short side size of the processed transform blocks of the color difference is not smaller than a predetermined threshold value, it is possible to be set to the value of the adaptive primary conversion flag luminance .
[0018]
　The inverse orthogonal transform unit, as the inverse orthogonal transform, is configured to perform the inverse primary horizontal transform is inverse primary transform in the horizontal direction, the inverse primary vertical conversion which is inverse primary transform in the vertical direction, the inverse primary horizontal conversion type of conversion, when the size of the horizontal width of the processed transform blocks of the color difference is greater than a predetermined threshold value, is set based on the horizontal transformation set and the primary horizontal conversion designation flag, convert the type of the inverse primary vertical transform , if the size of the vertical width of the processed transform blocks of the color difference is greater than a predetermined threshold value, it is possible to be set on the basis of the vertical transformation set and the primary vertical conversion designation flag.
[0019]
　Information relating to the inverse orthogonal transform, it is possible to include a conversion skip flag indicating whether to skip the inverse orthogonal transform processing.
[0020]
　The value of the conversion the skip flag of the color difference can be to be set to the value of the conversion the skip flag brightness.
[0021]
　Information relating to the inverse orthogonal transform, it is possible to include a secondary conversion identifier indicating whether to apply any inverse secondary transform.
[0022]
　The value of the secondary transform identifier of the color difference can be to be set to the value of the secondary transform identifier brightness.
[0023]
　The image processing method of the first embodiment of the present technology, by using the information about the inverse orthogonal transform of the color difference which is derived based on the information for the inverse orthogonal transform of the luminance, is an image processing method for performing the inverse orthogonal transformation of the color difference .
[0024]
　The image processing apparatus of the second embodiment of the present technology, the image processing and a orthogonal transform unit by using the information about the orthogonal transform of the color difference which is derived based on the information on the orthogonal transformation brightness, performs the orthogonal transform of the color difference it is a device.
[0025]
　The image processing method of the second embodiment of the present technology, using information on the orthogonal transformation of the color difference which is derived based on the information on the orthogonal transformation brightness, an image processing method for performing the orthogonal transform of color difference.
[0026]
　In the image processing apparatus and method of the first aspect of the present technique, the information is used for the inverse orthogonal transform of the color difference which is derived based on the information for the inverse orthogonal transform of the luminance, the inverse orthogonal transform of the color difference is performed.
[0027]
　In the image processing apparatus and method of the second aspect of the present technology is information is used regarding orthogonal transformation of the color difference which is derived based on the information on the orthogonal transformation brightness, the orthogonal transform of the color difference is performed.
Effect of the invention
[0028]
　According to the present disclosure, an image can be processed. In particular, it is possible to suppress the reduction of the coding efficiency.
BRIEF DESCRIPTION OF THE DRAWINGS
[0029]
FIG. 1 is a diagram showing the correspondence between the orthogonal transformation is selected as the transformation set.
The type of FIG. 2 orthogonal transformation is a diagram showing the correspondence between the function used.
Is a diagram illustrating a correspondence relationship [3] and transformation set as the prediction mode.
4 is a diagram illustrating an example of syntax and semantics of the conversion unit.
Is a diagram illustrating an example of FIG. 5] residual_coding syntax.
6 is a diagram showing an example of a parameter of the color difference deriving by using the value of luminance.
7 is a diagram illustrating an example of a method of deriving the adaptive primary conversion flag of the color difference.
[Figure 8] CU, diagrams PU, and TU of the shape will be described.
Is a block diagram showing a main configuration example of FIG. 9 the image decoding device.
It is a block diagram showing a main configuration example of FIG. 10 inverse transform unit.
11 is a block diagram showing a main configuration example of the color difference adaptive primary conversion information deriving unit.
Is a flowchart illustrating an example of the flow of FIG. 12 the image decoding processing.
13 is a flowchart illustrating an example of a flow of primary conversion information decoding process.
14 is a flowchart illustrating an example of the inverse conversion processing flow.
Is a flowchart illustrating an example of the flow of FIG. 15 the chrominance adaptive primary conversion information deriving process.
Is a flowchart illustrating an example of the flow of FIG. 16 inverse primary transform selection process.
17 is a flowchart describing an example of the flow of the color difference adaptive primary conversion information deriving process.
18 is a flowchart describing an example of the flow of the color difference adaptive primary conversion information deriving process.
It is a flowchart illustrating an example of FIG. 19 flow chrominance adaptive primary conversion information deriving process.
Diagrams [20] shows an example of syntax and semantics of the conversion unit.
21 is a diagram showing an example of a syntax Residual_coding.
Is a flowchart illustrating an example of FIG. 22 flow chrominance adaptive primary conversion information estimation flag decoding.
It is a flowchart illustrating an example of FIG. 23 flow chrominance adaptive primary conversion information deriving process.
Is a flowchart illustrating an example of the flow of FIG. 24 the chrominance adaptive primary conversion information deriving process.
[FIG 25 is a flowchart illustrating an example of a flow of primary horizontal conversion type deriving process.
FIG. 26 is a flowchart for explaining an example of the flow of the primary vertical conversion type deriving process.
FIG. 27 is a diagram illustrating an example of a method of deriving a transform skip flag of the color difference.
[FIG. 28] is a diagram illustrating an example of syntax Residual_coding.
FIG. 29 is a flowchart illustrating an example of a flow of conversion skip flag derivation process.
Is a diagram FIG. 30 shows an example of a syntax Residual_coding.
[FIG. 31] is a flow chart illustrating an example of a flow of conversion skip flag derivation process.
[FIG. 32] is a diagram showing an example of the syntax of Residual_coding.
[FIG 33 is a flowchart illustrating an example of a flow of conversion skip flag derivation process.
FIG. 34 is a diagram showing an example of a syntax Residual_coding.
[FIG. 35] is a flow chart illustrating an example of a flow of conversion skip flag derivation process.
[FIG. 36] is a diagram showing an example of the syntax and semantics of the conversion unit.
Is a diagram FIG. 37 shows an example of a syntax Residual_coding.
[FIG. 38] is a flow chart illustrating an example of a flow of conversion skip flag derivation process.
[39] is a diagram for explaining an example of a method of deriving the secondary conversion identifier of the color difference.
It is a diagram FIG. 40 shows an example of the syntax conversion unit.
[FIG. 41] is a diagram showing an example of the syntax of the conversion unit.
[FIG. 42] is a flowchart showing an example of a process for deriving the flow of secondary conversion identifier for chrominance.
[43] is a diagram showing an example of the syntax of the conversion unit.
[FIG. 44] is a flowchart showing an example of a process for deriving the flow of secondary conversion identifier for chrominance.
[FIG. 45] is a diagram illustrating an example of syntax of the conversion unit.
[FIG. 46] is a flowchart showing an example of a process for deriving the flow of secondary conversion identifier for chrominance.
[FIG. 47] is a diagram showing an example of the syntax of the conversion unit.
Is a flowchart illustrating an example of FIG. 48 the flow of process for deriving the secondary conversion identifier for chrominance.
[FIG. 49] is a diagram showing an example of the syntax of the conversion unit.
Is a flowchart illustrating an example of FIG. 50 the flow of process for deriving the secondary conversion identifier for chrominance.
[FIG. 51] is a block diagram showing a main configuration example of an image encoding apparatus.
[FIG. 52] is a block diagram showing a main configuration example of a conversion unit.
Is a block diagram showing a main configuration example of FIG. 53] color difference adaptive primary conversion information deriving unit.
[FIG. 54] is a flowchart illustrating an example of the flow of the image decoding processing.
FIG 55 is a flowchart illustrating an example of a conversion processing flow.
FIG 56 is a flowchart illustrating an example of a flow of primary conversion selection process.
FIG 57 is a flowchart illustrating an example of a flow of primary conversion information encoding process.
FIG 58 is a flowchart illustrating an example of a flow of the color difference adaptive primary conversion information estimation flag coding process.
FIG 59 is a flowchart illustrating an example of a conversion skip flag encoding processing flow.
FIG 60 is a flowchart illustrating an example of a conversion skip flag encoding processing flow.
FIG 61 is a flowchart illustrating an example of a conversion skip flag encoding processing flow.
FIG 62 is a flowchart illustrating an example of a conversion skip flag encoding processing flow.
FIG 63 is a flowchart illustrating an example of a conversion skip flag encoding processing flow.
FIG 64 is a flowchart illustrating an example of a flow for the chrominance secondary transform identifier encoding process.
Is a flowchart illustrating an example of the flow of FIG. 65] secondary transform identifier coding processing chrominance.
FIG 66 is a flowchart illustrating an example of a flow for the chrominance secondary transform identifier encoding process.
FIG 67 is a flowchart illustrating an example of a flow for the chrominance secondary transform identifier encoding process.
FIG 68 is a flowchart illustrating an example of a flow for the chrominance secondary transform identifier encoding process.
[FIG. 69] is a block diagram showing a main configuration example of a computer.
Is a block diagram showing an example of a schematic configuration of FIG. 70] television set.
[FIG. 71] is a block diagram showing an example of a schematic configuration of a mobile phone.
[FIG. 72] is a block diagram showing an example of a schematic configuration of a recording and reproducing apparatus.
[FIG. 73] is a block diagram showing an example of a schematic configuration of an imaging apparatus.
[FIG. 74] is a block diagram showing an example of a schematic configuration of a video set.
[FIG. 75] is a block diagram showing an example of a schematic configuration of the video processor.
[FIG. 76] is a block diagram showing another example of a schematic configuration of the video processor.
[FIG. 77] is a block diagram showing an example of a schematic configuration of a network system.
DESCRIPTION OF THE INVENTION
[0030]
　The following describes embodiments of the present disclosure (hereinafter referred to as embodiments). The description will be made in the following order.
　1. Orthogonal transformation of the color difference
　2. First Embodiment (image decoding apparatus, Apt_flag, Pt_idx)
　3. Second Embodiment (image decoding apparatus, Ts_flag)
　4. Third Embodiment (image decoding apparatus, St_idx)
　5. Fourth Embodiment (image coding apparatus, Apt_flag, Pt_idx)
　6. Fifth Embodiment (image coding apparatus, Ts_flag)
　7. Sixth Embodiment (image coding apparatus, St_idx)
　8. Otherwise
[0031]
　<1. Orthogonal transformation> of the color difference
　　
　In test model described in Non-Patent Document 1 (JEM4 (Joint Exploration Test Model 4)), for encoding efficiency in the high-resolution images such as 4K, CTU (Coding Tree Unit) the maximum size of the size is extended to 256x256 from 128x128. Further, as the structure of the block division, in addition to conventional quad-tree division introduces a binary tree of the horizontal / vertical directions, accordingly, in addition to the square of the transform blocks, rectangular conversion block is also introduced .
[0032]
　Further, the JEM4, of each transform block luminance, the primary conversion PThor (also referred to as primary horizontal conversion) in the horizontal direction and in each vertical direction of the primary conversion PTver (also referred to as primary vertical conversion) adaptively multiple different orthogonal from the conversion, adaptive primary transform to select the primary translation (AMT (Adaptive Multiple Core transforms)) is disclosed.
[0033]
　More specifically, the conversion block (also referred to as luminance conversion block) units of luminance, adaptive primary conversion flag apt_flag indicating whether or not to perform an adaptive primary conversion relating to the luminance (amt_flag, cu_pt_flag or also referred to as Emt_flag,) 0 for (false), the orthogonal transform to be applied to the horizontal primary conversion and vertical-primary conversion, DCT by (Discrete Cosine transform) -II or DST (Discrete Sine transform) -VII the mode information (uniquely) is determined.
[0034]
　In contrast, for example, if the adaptive primary conversion flag apt_flag regarding brightness 1 (true), as shown in Table (LUT_TrSetToTrTypIdx) shown in FIG. 1, each of the horizontal direction (x direction) and vertical direction (y-direction) transformation set TrSet comprising a candidate to become orthogonal transformation of the primary transformation for is selected from among four candidate (transform set Idx = 0 to 3). DST-VII and DCT-VIII or the like shown in FIG. 1 shows a type of orthogonal transform, respectively, functions as shown in the table of FIG. 2 is used.
[0035]
　In Table 2, the identifier corresponding to each type of the orthogonal transform (transform type identifier TrTypeIdx) is defined. For example, if the orthogonal transform type (Transform Type) is DCT-II, the value of the conversion type identifier TrTypeIdx 0 is assigned.
[0036]
　Selection of transformation set TransformSet (decision) will vary by the prediction type of CU conversion block to be processed belongs (Coding Unit) (also referred to as coding units). For example, the prediction type is if intra prediction, as shown in Table (LUT_IntraModeToTrSet) shown in FIG. 3, is performed on the basis of the intra prediction mode (IntraPredMode). For example, implemented to set the following formula (1) and as shown in equation (2), conversion set identifier TrSetIdx be specified for each direction transformation set (TrSetH, TrSetV), the corresponding transformation set TrSet that.
[0037]
　TrSetH = LUT_IntraModeToTrSet [ IntraPredMode ] [ H(=0) ]
　・・・（１）
　TrSetV = LUT_IntraModeToTrSet [ IntraPredMode ] [ V(=1) ]
　・・・（２）
[0038]
　Here, TrSetH represents the transformation set of primary horizontal conversion PThor (also referred to as the primary horizontal transformation set), TrSetV shows transformation set of primary vertical conversion PTver (also referred to as primary to Chok transformation set). Also, the look-up table LUT_IntraModeToTrSet shows the correspondence table of FIG. Lookup table LUT_IntraModeToTrSet [] [] 1 th sequence of the intra prediction mode IntraPredMode as arguments, the second array, and argument {H = 0, V = 1}.
[0039]
　For example, in the case of the intra prediction mode number 18 (IntraPredMode == 18), as the primary horizontal transformation set TrSetH, transformation set of transformation set identifier TrSetIdx = 2 shown in Table (LUT_TrSetToTrTypeIdx) of FIG. 1 is selected, the primary vertical conversion set as TrSetV, transformation set of transformation set identifier TrSetIdx = 0 shown in the table of FIG. 1 is selected.
[0040]
　If the prediction type is an inter-prediction, the direction of the transformation set (TrSetH, and TrSetV) relative conversion set identifier TrSetIdx (= InterTrSetIdx) the following expression that specifies the transformation set TrSet for inter prediction (3) Ya is set as in equation (4). For example, if the table of FIG. 1, the value of InterTrSet is 3.
[0041]
　TrSetH = InterTrSetIdx
　・・・（３）
　TrSetV = InterTrSetIdx
　・・・（４）
[0042]
　Furthermore, to apply any orthogonal transformation of the transformation set TrSet chosen primary horizontal conversion is selected by the primary horizontal conversion designation flag Pt_hor_flag. Also, apply any orthogonal transformation of the transformation set TrSet chosen primary vertical conversion is selected by the primary vertical conversion designation flag Pt_ver_flag. For example, as shown in the following expression (5) and (6), primary {horizontal, vertical} transformation set Trset {H, V}, a primary {horizontal, vertical} conversion specification flag pt_ {hor, ver} the _flag as an argument, it is derived from the definition table of conversion set shown in FIG. 1 (LUT_TrSetToTrTypeIdx).
[0043]
　TrTypeIdxH = LUT_TrSetToTrTypeIdx [TrSetH] [pt_hor_flag]
　· · · (5)
　TrTypeIdxV = LUT_TrSetToTrTypeIdx [TrSetV] [pt_ver_flag]
　· · · (6)
[0044]
　For example, in the case of the intra prediction mode number 18 (IntraPredMode == 18), from the table of FIG. 3, the value of the transformation set identifier TrSetIdx primary horizontal transformation set TrSetH is 2, on the transformation set definition table LUT_TrSetToTrTypeIdx in FIG orthogonal transform to be applied is selected (specified) from the conversion set identifier TrSetIdx == 2 a is converted set into a primary horizontal translation. That is, if the primary horizontal conversion designation flag pt_hor_flag is 0, the horizontal conversion type identifier TrTypeIdxH that specifies the type of orthogonal transform of the primary horizontal conversion PThor, as shown in Table 1, the conversion type identifier TrTypeIdx showing a DST-VII is set to the value "4", if the primary horizontal conversion designation flag pt_hor_flag is 1, the horizontal conversion type identifier TrTypeIdxH, the value of the conversion type identifier TrTypeIdx showing a DCT-V "1" is set.
[0045]
　Incidentally, the primary conversion identifier pt_idx from the primary horizontal conversion designation flag pt_hor_flag and primary vertical conversion designation flag pt_ver_flag is derived based on the following equation (7).
[0046]
　pt_idx = (pt_ver_flag << 1) + pt_hor_flag
　・・・（７）
[0047]
　That is, the upper 1bit primary conversion identifier pt_idx corresponds to the value of the primary vertical conversion designation flag, lower 1bit corresponds to the value of the primary horizontal conversion designation flag. Respect derived bin string in the primary conversion identifier Pt_idx, by applying arithmetic coding, to generate a bit string, the encoding is performed. Incidentally, Pt_idx the upper 1bit, primary horizontal conversion specifying flag may be set backwards 1bit as the primary vertical conversion designation flag.
[0048]
　In contrast, for the conversion block of the color difference (Cb, Cr), adaptive-primary conversion is not applied, the type of orthogonal transform primary horizontal transform and the primary vertical conversion, DCT-II is always selected.
[0049]
　A in FIG. 4 is an example of a syntax table translation unit TU, B of FIG. 4 shows an example of semantics corresponding thereto. Residual data existence flag cbf sign SYN11 is shown in row attached in the syntax shown in A of FIG. 4 [x0] [y0] [compID] (coded_block_flag), as shown in B of FIG. 4, non-zero coefficients in the transform block of the color signal is specified by the color signal identifier compID is a flag indicating whether or not there are one or more. If the flag is 1 (true), a non-zero coefficient in the corresponding transformation blocks indicates that there are one or more, if the flag is 0 (false), the non-zero coefficient in the corresponding transformation blocks is not present It is shown. For convenience, the residual data existence flag cbf luminance [x0] [y0] [COMPONENT_Y] a may be referred to as cbf_luma (cbf_luma = cbf [x0] [y0] [COMPONENT_Y]).
[0050]
　The adaptive-primary conversion enable flag apt_enabled_flag shown in A in FIG. 4, as shown in B of FIG. 4, which is information about the permission of the adaptive-primary conversion. If the value is 1 (true), on the encoded data, indicating that information on adaptation primary transform may be present. Also, if this value is 0 (false), on the encoded data, indicating information on adaptive primary translation does not exist.
[0051]
　The adaptive primary conversion flag apt_flag relating to the luminance conversion block shown in line code SYN12 is affixed in the syntax shown in A in FIG. 4, as shown in B of FIG. 4, the converted block of the luminance, it is a flag indicating whether to apply adaptive primary conversion. If the value is 1 (true), and applying an adaptive-primary conversion, the case of 0 (false), do not apply the applicable primary conversion. As shown in the syntax table of A in FIG. 4, the residual data existence flag cbf_luma luminance is "1" (i.e. true), and if the adaptive primary conversion enable flag apt_enabled_flag is "1" (i.e., true) , adaptive primary conversion flag apt_flag luminance is coded (decoded). If there is no apt_flag on encoded data is interpreted as a value of apt_flag is 0.
[0052]
　Further, in the syntax shown in A of FIG. 4, the residual data residual_coding transform block of the color signal is specified by the color signal identifier compID shown in row code SYN13 is attached () is the corresponding color signal remaining If the difference data existence flag is 1 (true) is encoded (decoded).
[0053]
　Figure 5 shows an example of the syntax table Residual_coding. In the syntax table shown in FIG. 5, the primary conversion identifier pt_idx the luminance indicated by reference numeral SYN22, as shown in the syntax table, adaptive primary conversion flag apt_flag luminance is 1 (true), transform quantization bypass a flag transquant_bypass_flag is 0 (false), a conversion skip flag ts_flag is 0 (false), even when the color signal identifier compID represents luminance (compID == COMPONENT_Y), non-zero present in the transform block the total number of coefficients numSig is equal to or greater than the predetermined threshold value PtNumSigTH, the long side of the transform block (max (log2TBWSize, log2TBHSize)) is in the case of less than a predetermined threshold value MaxPTSize, is encoded (decoded). If there is no primary conversion identifier pt_idx luminance on the encoded data is interpreted as a value of pt_idx is 0.
[0054]
　In Non-Patent Document 1, only the luminance (Y), although the adaptive-primary conversion can be applied, not applied to the color difference (Cb, Cr). Therefore, coding efficiency of the primary converter for the color difference, there is a risk of reduced compared to luminance.
[0055]
　In contrast, for example, similarly to the luminance (Y), color difference (Cb, Cr) for each transform block of the adaptive primary conversion flag Apt_flag, it is conceivable to explicitly encode the primary transform identifier Pt_idx. However, in this method, for encoding the adaptive primary conversion flag apt_flag and primary transform identifier pt_idx regarding the color difference, the code amount increases, the coding efficiency is likely to be reduced.
[0056]
　Incidentally, is added that, in this case, in the encoding side, the color signal (Y, Cb, Cr) whether the selection of the adaptive primary conversion for each, and must be a mode decision regarding the determination of the primary conversion identifier, processing there is a possibility that the amount is increased. Also in the decoding side, the color signal (Y, Cb, Cr) must decode adaptive primary conversion flag apt_flag and primary transform identifier pt_idx each, there is a possibility that the processing amount increases.
[0057]
　　
　Accordingly, so as to derive on the basis of information about the (inverse) orthogonal conversion of the color difference information on (inverse) orthogonal transformation brightness. That is, using information about the (inverse) orthogonal transformation of the color difference which is derived based on the information about the (inverse) orthogonal transformation brightness, to perform the (inverse) orthogonal conversion of the color difference. For example, in the image processing apparatus, using information on (inverse) orthogonal transformation of the color difference which is derived based on the information about the (inverse) orthogonal transformation brightness, performing (inverse) orthogonal conversion of the color difference (inverse) orthogonal transform unit the to be provided.
[0058]
　In this way, it is possible to omit the information coding and decoding of related (inverse) orthogonal conversion of the color difference, suppressing an increase in the amount of codes, it is possible to suppress the reduction of the coding efficiency. Further, it is possible to suppress increase in the load of the encoding and decoding. In this specification, the orthogonal transform and inverse orthogonal transform is an inverse process to each other, for example by inverse orthogonal transformation on the orthogonal transformed data, it shall be able to recover the data before orthogonal transform. Moreover, information on the orthogonal transformation is information utilized for orthogonal transform, it may possible be used for inverse orthogonal transform. The information for the inverse orthogonal transformation is information to be used for inverse orthogonal transform, it may be possible to use the orthogonal transform. That is, in this specification, the information about the information and inverse orthogonal transform about orthogonal transformation (may include the same information with each other) also possible to refer to the same information to each other. Above it the primary transform and inverse primary transform is the same between the secondary and inverse secondary transform.
[0059]
　Content of the information about the (inverse) orthogonal transform is arbitrary. For example, as shown in the table of FIG. 6, in the conversion block to be processed, adapted for use as an adaptively selected and (reverse) primary transform one from among a plurality of different inverse orthogonal transform (inverse) primary transform it may include an adaptive primary conversion flag apt_flag indicating whether to apply. It is also possible to include a primary conversion identifier pt_idx indicating whether to apply any (inverse) primary transform in (inverse) primary transform in the vertical direction and the horizontal direction.
[0060]
　For example, the color difference and (Cb) adaptive primary conversion flag apt_flag [Cb] color difference (Cr) adaptive primary conversion flag apt_flag [Cr] and the respective, by deriving based on adaptive primary conversion flag apt_flag [Y] of the luminance , it is possible to suppress the reduction of the coding efficiency of the color difference (Cb / Cr). Further, it is possible to suppress the increase in overhead associated code amount. Further, for example, color difference primary conversion identifier Pt_idx of (Cb) [Cb] and color difference (Cr) of the primary converter identifier pt_idx [Cr] and respectively, by deriving, based on the primary conversion identifier pt_idx [Y] of the luminance, it is possible to suppress the reduction of the coding efficiency of the color difference (Cb / Cr). Further, it is possible to suppress the increase in overhead associated code amount. Incidentally, the primary conversion identifiers color difference (may be shared) may be a common and Cb and Cr.
[0061]
　Further described with reference to the table of FIG. Conventionally, No. As in the row of # 0, adaptation (inverse) primary transform is not employed in the color difference adaptive primary conversion flag apt_flag and primary transform identifier pt_idx color difference was always omitted. In this case, the disuse of the adaptive (inverse) primary transform, coding efficiency is likely to be reduced as described above.
[0062]
　In contrast, No. # As one line, also can be applied to adaptation (inverse) primary transform in the color difference, further always an adaptive primary conversion flag apt_flag and primary transform identifier pt_idx chrominance, luminance adaptation so as to estimate the primary conversion flag apt_flag and primary transform identifier Pt_idx. For example, the value of the adaptive primary conversion flag apt_flag [Cb / Cr] of the color difference may be set the value of the luminance of the adaptive primary conversion flag apt_flag [Y]. Further, for example, adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] is set to the value of true if the primary conversion of the color difference identifier pt_idx [Cb / Cr] value in primary conversion identifier luminance Pt_idx of [Y], If adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] is false, it may be set to a predetermined value to the value of the primary conversion identifier pt_idx [Cb / Cr] of the color difference. In this way omitted, not only it can be applied also adaptation (inverse) primary transform to the color difference, the transmission of the adaptive primary conversion flag apt_flag and primary transform identifier pt_idx color difference (coding and decoding) it is possible, it is possible to suppress the reduction of the coding efficiency. Further, it is possible to suppress increase in the load of the encoding and decoding.
[0063]
　Also, No. # As second row, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, the prediction type of the coding block transform block to be processed belongs is inter prediction (CuPredMode == If it is MODE_INTER), it sets the value of the luminance of the adaptive primary conversion flag apt_flag [Y] to the value of the adaptive primary conversion flag Apt_flag of the color difference [Cb / Cr], when the prediction type of the coded block is an intra prediction , the value of the adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] may be set to 0 (false). By doing so, not only can be applied also adaptation (inverse) primary transform to the color difference, when the inter prediction modes tendency of the residual signal are similar only, the value of the adaptive primary conversion flag luminance it is possible to make use of, it is possible to suppress the reduction of the coding efficiency.
[0064]
　Further, No. # As the third row, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, when the prediction type of the coding block transform block to be processed belongs is an inter prediction, or, if an intra prediction is prediction mode in the luminance and chrominance match, sets the value of the luminance of the adaptive primary conversion flag apt_flag [Y] to the value of the adaptive primary conversion flag Apt_flag of the color difference [Cb / Cr], the If the prediction type of the coding block is an intra prediction does not match the prediction mode and the luminance and color difference, the value of the adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] may be set to 0 (false) . By doing so, not only can be applied also adaptation (inverse) primary transform to the color difference, when the prediction modes tendency of the residual signal are similar only, the value of the adaptive primary conversion flag luminance it is possible to make use, it is possible to suppress the reduction of the coding efficiency.
[0065]
　Also, No. # As 4 rows, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, when the prediction type of the coding block transform block to be processed belongs is an inter prediction, or, if the prediction mode is the intra prediction of an intra block copy, set the value of the luminance of the adaptive primary conversion flag apt_flag [Y] to the value of the adaptive primary conversion flag Apt_flag of the color difference [Cb / Cr], the coded block the prediction type is a prediction mode is an intra block copy when an intra prediction not (IntraBC, intra block copy, referred to as intra motion compensation), the value of the adaptive primary conversion flag apt_flag [Cb / Cr] chrominance 0 (false ) may be set. By doing so, not only can be applied also adaptation (inverse) primary transform to the color difference, when the prediction modes tendency of the residual signal are similar only, the value of the adaptive primary conversion flag luminance it is possible to make use, it is possible to suppress the reduction of the coding efficiency.
[0066]
　Also, No. # As in the fifth row, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, adaptive primary conversion of the color difference flag apt_flag [Cb / Cr] an adaptive primary conversion flag luminance Apt_flag If [Y] color difference adaptive primary conversion information estimation flag indicating whether the estimated based on chroma_apt_info_infer_flag is 1 (true), the value luminance adaptive primary conversion to the adaptive primary conversion flag Apt_flag of the color difference [Cb / Cr] flag Apt_flag [ sets the value of Y], when the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag is 0 (false), be set to 0 (false) to the value of the adaptive primary conversion flag apt_flag [Cb / Cr] color difference good. In this way, it is possible whether to use the value of the adaptive primary conversion flag luminance explicitly controlled by using the color difference adaptive primary conversion information estimation flag Chroma_apt_info_infer_flag. Thus, readily, only if a sufficiently large effect can be obtained since it is possible to so as to use the value of the adaptive primary conversion flag of luminance, it is possible to further suppress the reduction of the coding efficiency.
[0067]
　Also, No. As rows of # 6, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, when the short side size of the transform block processed in the color difference is greater than a predetermined threshold value, when set to the value of the adaptive primary conversion flag apt_flag [Cb / Cr] value luminance adaptive primary conversion flag of the Apt_flag of [Y] of the color difference, the size of the short sides of the conversion block is less than the threshold, the color difference adaptive primary the value of the conversion flag apt_flag [Cb / Cr] may be set to 0 (false). Since the by, it is possible to avoid applying adaptive (inverse) primary conversion for small transform block is not sufficiently large effect is obtained, it is possible to suppress the reduction of the coding efficiency together, the adaptive (inverse) primary transformation, it is possible to suppress an increase in the unnecessary circuit scale by applying the smaller transform blocks not sufficiently large effect can be obtained.
[0068]
　Also, No. As in the row of # 7, also can be applied to adaptation (inverse) primary transform in the color difference, further, for example, when the size of the horizontal width of the processed transform blocks of the color difference is less than a predetermined threshold value, ( the conversion type reverse) primary horizontal conversion is set to a predetermined conversion type, when the size of the horizontal width of the transform block is greater than the threshold, (reverse) horizontal transformation set and the primary horizontal conversion specifies the conversion type primary horizontal conversion it may be set based on the flag. Similarly, if the size of the vertical width of the processed transform blocks of the color difference is less than a predetermined threshold value, the (inverse) to set the conversion type primary vertical conversion into a predetermined conversion type, the size of the vertical width of the transform block If greater than the threshold may be set based on the conversion type (reverse) primary vertical conversion in the vertical transformation set and the primary vertical conversion designation flag. By doing so, (relative to the width direction (inverse) primary transform) sufficiently large effect as can not be obtained width relative narrow transform block, so as not to apply the adaptive (inverse) primary transform it is possible to, it is possible to further suppress the reduction of the coding efficiency, unnecessary by applying this adaptation (inverse) primary conversion, the narrow transform block enough not sufficiently large effect can be obtained it is possible to suppress an increase in circuit size.
[0069]
　It is also possible to combine any of the cases described above. For example, No. As in the row of # 8, No. # And 2 rows of the case, No. # May be combined with the 7 rows of the case. In this way, it is possible to obtain the effect obtained in each case. Moreover, each case described above, can also be combined with other cases which are not described above. For example, No. In # 2 rows of the case, when the prediction type of the encoding block to be processed is intra prediction, and transmits the adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] (signaling) manner it may be. That is, in this case, encoding and decoding of the adaptive primary conversion flag apt_flag of the color difference [Cb / Cr] is performed. When the value of the adaptive primary conversion flag apt_flag [Cb / Cr] color difference is 1 (true), and further, transmits the adaptive primary conversion identifier pt_idx of the color difference [Cb / Cr] may be (signaling to) as . That is, in this case, encoding and decoding of the adaptive primary conversion identifier pt_idx of the color difference [Cb / Cr] is performed.
[0070]
　　
　wherein the bit stream is a portion of a moving image in (encoded data) area (processing unit) CU (Coding Unit), PU (Prediction Unit), and the shape of the TU explaining . CU shown in FIG. 8, PU, and TU (Transform Unit) is, JVET-C0024, "EE2.1: Quadtree plus binary tree structure integration with JEM tools" are described in QTBT of (Quad tree plus binary tree) CU, PU, and a TU. Specifically, the block division of the CU, one block 4 (= 2x2) pieces as well, 2 (= 1x2,2x1) can also be divided into sub-blocks. That is, the block division of the CU is performed by repeating the division of one block to the four or two sub-blocks recursively, resulting quadtree (Quad-Tree) shape or horizontal or vertical direction Binary tree (Binary-tree) like a tree structure is formed of.
[0071]
　As a result, the shape of the CU, square well, it may be rectangular. For example, LCU (Largest Coding Unit) If the size is 128x128, (size h of size w × vertical horizontal) size CU, as shown in FIG. 8, 128X128,64x64,32x32,16x16,8x8 , not only the size of the square, such as 4x4, 128x64,128x32,128x16,128x8,128x4,64x128,32x128,16x128,8x128,4x128,64x32,64x16,64x8,64x4,32x64,16x64,8x64,4x64,32x16,32x8, 32X4,16x32,8x32,4x32,16x8,16x4,8x16,4x16, might be a rectangle the size of such 8X4,4x8. In addition, PU and TU is the same as the CU.
[0072]
　Further, the TU, include conversion block, and converts the block of the color difference (Cb / Cr) of the luminance (Y). Color format is 4: 2: 0 (e.g., YUV420) For the ratio of the picture size of the color difference with respect to the picture size of the luminance is half the vertical width and the horizontal width of both. Thus, if the transform block size of the luminance 8x4, transform block size of the corresponding color difference becomes 4x2. The color format is 4: 2: 2 (e.g., YUV422) For the ratio of the picture size of the color difference with respect to the picture size of the luminance vertical width is 1/2, the width is 1. Thus, if the transform block size of the luminance 8x4, transform block size of the corresponding color difference becomes 8x2. The color format is 4: 4: 4 (e.g., YUV 444) if the ratio of the picture size of the color difference with respect to the picture size of the luminance vertical width of 1, the width is 1. Thus, if the transform block size of the luminance 8x4, transform block size of the corresponding color difference is 8x4.
[0073]
　Regarding the I-slice may be coded as separate CU out with the luminance (Y) and color difference (Cb / Cr). In this case, it has the effect that in the luminance and color difference, it is possible to take a split structure of different CU, improves the coding efficiency of I-slices. In the following, for convenience, in the same CU, it is described as being included luminance and chrominance information, but is not limited thereto.
[0074]
　<2. First Embodiment>
　　
　FIG. 9 is a block diagram showing an example of a configuration of an image decoding apparatus which is an embodiment of an image processing apparatus to which the present technology is applied. The image decoding apparatus 100 shown in FIG. 9, as in the AVC and HEVC, prediction residual image and its prediction picture is a device for decoding a coded data. For example, the image decoding apparatus 100 includes technologies and proposed in HEVC, implementing the techniques proposed by JVET (Joint Video Exploration Team).
[0075]
　9, the image decoding apparatus 100, decoding unit 111, an inverse quantization unit 112, inverse transform unit 113, calculation unit 114, a frame memory 115 and the prediction unit 116,. Incidentally, the prediction unit 116 includes an intra prediction unit (not shown), and an inter prediction unit. The image decoding apparatus 100, by decoding the coded data # 1 (bitstream), an apparatus for generating a moving image # 2.
[0076]
　Decoding section 111 receives as input encoded data # 1, along the defined syntax table, a bit string of coded data # 1, to the variable-length decoding the syntax values ​​of each syntax element. Further, the syntax elements, header information Hinfo, prediction mode information pinfo, conversion information TINFO, includes information such as the residual information rinfo.
[0077]
　The header information Hinfo such VPS / SPS / PPS / slice header SH, the image size (width PicWidth, longitudinal width PicHeight), bit depth (brightness BitDepthY, chrominance bitDepthC), the maximum value of the CU size MaxCUSize / minimum MinCUSize, 4 min tree partitioning maximum depth MaxBTDepth / minimum depth MinBTDepth, maximum MaxTSSize (maximum conversion skipping conversion skip blocks of maximum depth MaxQTDepth / minimum depth MinQTDepth, 2-tree division (also referred to as Quad-tree division) (Binary-tree division) also referred to as block size), also referred to as on-off flag (valid flag for each coding tool) includes information defining the like.
[0078]
　For example, the on-off flag of the encoding tools included in the header information Hinfo, conversion shown below, there is on-off flag related to the quantization process. Incidentally, off flag of the encoding tools can syntax relating to the encoded tool is interpreted as a flag indicating whether or not present in the encoded data. When the value of the on-off flag is 1 (true) indicates that the coded tool is available, if the value of the on-off flag is 0 (false), that the coded tool is not available show. Note that the interpretation of the flag values ​​may be reversed.
[0079]
　Adaptive-primary conversion enable flag apt_enabled_flag (adaptive_primary_transform_enabled_flag, adaptive_pt_enabled_flag or also referred to as Amt_enabled_flag,) as one of the conversion process and its inverse processing, adaptive primary conversion Choices coding tools (also referred to as adaptive primary transform) a flag indicating whether available.
[0080]
　Secondary transformation valid flag st_enabled_flag as one of the conversion process and its inverse processing, coding tool for a secondary transformation / inverse secondary transform is a flag indicating whether available.
[0081]
　Transformed and quantized bypass valid flag transquant_bypass_enabled_flag is transform and quantization, and as one of the reverse process, the encoding tool to skip an inverse transform transformation and quantization / inverse quantization is a flag indicating whether available is there.
[0082]
　Converting skip flag validity flag ts_enabled_flag as one of the conversion process and its inverse processing, two-dimensional transform skip and one-dimensional transform skip a flag indicating whether available. Part 2 dimensional transform skip is an encoding tool to skip the orthogonal transform and its inverse processing including the primary conversion and secondary conversion (inverse orthogonal transform). Also, one-dimensional transform skip among the primary conversion, the primary conversion of the horizontal or vertical direction, and inverse transform (inverse primary transform) corresponding thereto, as well as the encoding tool to skip the secondary conversion and its inverse secondary transform it is.
[0083]
　The prediction mode information pinfo, further, PU size to be processed PU (prediction block size) PUSize, intra prediction mode information IPinfo (e.g., JCTVC-W1005, 7.3.8.5 Coding Unit syntax in prev_intra_luma_pred_flag, mpm_idx, rem_intra_pred_mode, the thin luminance intra prediction mode IntraPredModeY derived from tax, chrominance intra-prediction mode IntraPredModeC etc.), motion prediction information MVinfo (e.g., JCTVC-W1005, see 7.3.8.6 prediction Unit Syntax, merge_idx, merge_flag, inter_pred_idc, ref_idx_LX, mvp_lX_flag, X = {0,1}, contained mvd) and the like.
[0084]
　The conversion into the information TINFO, for example, the width size TBWidth and vertical width TBHeight of processed transform block, (inverse) transformation and (inverse) transform quantization bypass flag transquant_bypass_flag indicating whether to skip the process of quantization, 2D transformation skip conversion skip flag indicating whether to apply Ts_flag, target conversion block adaptive primary conversion flag indicating whether to apply adaptive primary conversion in Apt_flag, which in the vertical and horizontal direction (reverse) primary transformation ( inverse) primary transform identifier pt_idx indicating whether to apply the primary conversion, which (inverse) secondary transformation indicates whether to apply the secondary transform identifier st_idx (dnsst_idx, also referred to as nsst_idx or Rot_idx,), scan identifier ScanIdx, quantization parameter qp include syntax, such as quantization matrices Scaling_matrix. Instead of the width size TBWidth and vertical width TBHeight of processed transform block, each base two TBWidth, logarithmic value of TBHeight log2TBWidth (log2TBWSize, also referred to as Log2TBW), include log2TBHeight (log2TBHSize, also referred to as Log2TBH) it may be so.
[0085]
　The residual information rinfo, for example, the residual data existence flag (cbf (coded_block_flag)), the last non-zero coefficient X-coordinate (last_sig_coeff_x_pos), last non-zero coefficient Y coordinates (last_sig_coeff_y_pos), sub-block non-zero coefficient presence flag (Coded_sub_block_flag ), non-zero coefficient presence flag (Sig_coeff_flag), a flag indicating the level or greater than 1 non-zero coefficients (gr1_flag) (also GR1 flag referred to), flag the level of non-zero coefficients indicate greater than 2 (Gr2_flag) (GR2 also flag referred to), (also referred to as sign code) code (Sign_flag) indicating the sign of non-zero coefficients, as well, so that such a non-zero coefficient residual level (Coeff_abs_level_remaining) (also referred to as non-zero coefficients residual level) is included it may be.
[0086]
　Decoding unit 111 refers to the residual information rinfo, to derive the quantized transform coefficient levels level for each coefficient position in each transform block. Decoding unit 111 supplies the prediction mode information Pinfo obtained by decoding the quantized transform coefficient levels level, the conversion information Tinfo to each block. For example, the decoding unit 111 supplies the prediction mode information Pinfo the inverse transform unit 113 and the prediction unit 116, the quantized transform coefficient levels level, and supplied to the inverse quantization unit 112, a conversion information TINFO, inverse transform unit 113 and it supplies the inverse quantization unit 112.
[0087]
　Inverse quantization unit 112, the conversion information TINFO, and inputs the quantized transform coefficient levels level, based on the conversion information TINFO, scaling the value of the quantized transform coefficient levels level (inverse quantization) and, after dequantization and outputs the conversion coefficient Coeff_IQ to the inverse transform unit 113.
[0088]
　Inverse transform unit 113, transform coefficient Coeff_IQ, conversion information TINFO, as input prediction mode information pinfo, based on the conversion information TINFO and prediction mode information pinfo like, the converted coefficient Coeff_IQ, applying an inverse transform, prediction deriving a difference D ', and outputs it to the arithmetic unit 114. Incidentally, the inverse transform is the inverse process of the conversion process performed in the encoding side, or the like. For example, the inverse orthogonal transform such as an inverse processing of orthogonal transform performed in the encoding side and the like are included in the inverse transform. Also, for example, if the primary conversion or secondary translation is performed as the orthogonal transform in the encoding side, and the inverse primary transform is an inverse processing of the primary translation, inverse secondary transform or the like which is the reverse processing of the secondary transform is included in the inverse transform . For details of the inverse transform unit 113 will be described later.
[0089]
　Calculation unit 114, corresponding to the prediction residual D 'and inputs the prediction image P supplied from the prediction unit 116, the following equation (8) prediction residual D as shown in' a prediction residual D ' It adds the predictive picture P (prediction signal) to derive the local decoded image Rec, supplies it to the outside of the frame memory 115 or the image decoding apparatus 100.
[0090]
　Rec = D' + P
　・・・（８）
[0091]
　The frame memory 115 receives the local decoded image Rec supplied from the arithmetic unit 114, and reconstructing a decoded image for each picture unit, and stored in the buffer in the frame memory 115. The frame memory 115 supplies from the buffer the decoded image is designated by the prediction mode information Pinfo the prediction unit 116 as a reference image read, to the prediction unit 116. The frame memory 115, the header information Hinfo according to generation of the decoded image, the prediction mode information pinfo, may be such as conversion information Tinfo to be stored in the buffer in the frame memory.
[0092]
　Prediction unit 116 inputs the prediction mode information pinfo, the prediction method specified by the prediction mode information pinfo, predicted picture P is designated by the prediction mode information pinfo, the decoded image stored in the frame memory 115 as a reference picture and outputs it to the arithmetic unit 114.
[0093]
　　
　FIG. 10 is a block diagram showing a main configuration example of the inverse conversion unit 113 included in the image decoding apparatus 100 of FIG. 9. As shown in FIG. 9, the inverse transform unit 113, a switch 121, the inverse secondary transform unit 122, and the inverse primary transform unit 123.
[0094]
　Switch 121 converts coefficients Coeff_IQ, and converts the skip flag ts_flag [compID] the input. If the value of the conversion the skip flag ts_flag [compID] is NO_TS (= 0), i.e., if you do not apply transform skip, switch 121 outputs the conversion coefficient Coeff_IQ to the inverse secondary transform unit 122. When the value of the conversion the skip flag ts_flag [compID] is 2D_TS (= 1), i.e., to indicate that the application of two-dimensional transform skip, switch 121, the inverse secondary transform unit 122 and the inverse primary transform unit 123 skip, and outputs a transform coefficient Coeff_IQ as prediction residual D '.
[0095]
　Inverse secondary transform unit 122 performs processing for the inverse secondary transform that is the inverse processing of the secondary transform is a predetermined conversion process, for example, orthogonal transform such. For example, the inverse secondary transform unit 122, a secondary conversion identifier St_idx, scan identifier scanIdx to scan a transform coefficient, and the transform coefficients Coeff_IQ as input, also referred to as referred to as transform coefficients Coeff_IS (primary transform coefficients Coeff_P after the inverse secondary transform ) derives and supplies the inverse primary transform unit 123. Conversely More specifically, the secondary transform identifier St_idx is, to indicate that applying an inverse secondary transform (st_idx> 0), the inverse secondary transform unit 122, the relative transform coefficients Coeff_IQ, corresponding to the secondary conversion identifier St_idx executing the processing of secondary conversion, and outputs the transform coefficients Coeff_IS after the inverse secondary transform. Secondary transformation identifier St_idx is, the case shown not to apply the inverse secondary transform (st_idx == 0), the inverse secondary transform unit 122 skips the inverse secondary transform, the transform coefficients Coeff_IQ as transform coefficients Coeff_IS after the inverse secondary transform Output.
[0096]
　Inverse primary transform unit 123 performs processing for the inverse primary transform is an inverse processing of the primary translation is a predetermined conversion process, for example orthogonal transformation or the like. For example, the inverse primary transform unit 123, the color signal identifier compID, the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y], the primary translation identifier pt_idx luminance [COMPONENT_Y], prediction mode information pinfo, a pair of size (width of the transform block numerical log2TBWSize , the logarithm Log2TBHSize) and conversion coefficient Coeff_IS after the inverse secondary transform the vertical width and the input. Inverse primary transform unit 123, the prediction mode information pinfo, color signals identifier compID, adaptive primary conversion flag apt_flag luminance [COMPONENT_Y], the primary conversion identifier pt_idx luminance [COMPONENT_Y], and references the primary conversion identifier pt_idx luminance [COMPONENT_Y] to, select the conversion type TrTypeIdxH, and conversion type TrTypeIdxV inverse primary vertical conversion of the inverse primary horizontal conversion of the color signal specified by the color signal identifier compID, the converted coefficient Coeff_IS after the inverse secondary reverse primary vertical inverse primary vertical conversion determined by the vertical width log2TBHSize conversion type TrTypeIdxV the conversion block, and the inverse primary horizontal transformation determined by the width log2TBWSize inverse primary horizontal transform type TrTypeIdxH and conversion block, the prediction residual D after the inverse primary transform ' derived, and outputs.
[0097]
　　
　As shown in Figure 11, the inverse primary transform unit 123, the color difference adaptive primary conversion information deriving unit 131, the inverse primary transform selection portion 132, the inverse primary vertical conversion unit 133, and the inverse primary horizontal transform unit with a 134.
[0098]
　Color difference adaptive primary conversion information deriving unit 131, when the color signal identifier compID indicates the color difference (compID == COMPONENT_Cb, COMPONENT_Cr), adaptive-primary conversion information of the luminance conversion block corresponding to the color difference conversion block (apt_flag [COMPONENT_Y], pt_idx [ COMPONENT_Y]) using adaptive primary conversion information of the color difference (apt_flag [compID], pt_idx [compID] (compID = COMPONENT_Cb, COMPONENT_Cr) derives), and supplies the result to the inverse primary transform selection portion 132.
[0099]
　　
　FIG. 11 is a functional block diagram showing a main configuration example of a function of the color difference adaptive primary conversion information deriving unit 131 has. As shown in FIG. 11, the color difference adaptive primary conversion information deriving unit 131 has a apt_flag deriving unit 151 and the pt_idx deriving unit 152. apt_flag deriving unit 151 performs processing relating to the derivation of the adaptive primary conversion flag apt_flag color difference.
[0100]
　For example, Apt_flag deriving unit 151, adaptive primary conversion flag Apt_flag regarding the color difference conversion block to be processed [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and adaptive primary conversion flag for the corresponding luminance conversion block into transform blocks of the color difference Apt_flag [ based on COMPONENT_Y], is derived as the following equation (9).
[0101]
　apt_flag[compID] = apt_flag[COMPONENT_Y]
　（ただし、compID=COMPONENT_Cb, COMPONENT_Cr）
　・・・（９）
[0102]
　That, Apt_flag deriving section 151, adaptive primary conversion flag Apt_flag chrominance [compID] (compID = COMPONENT_Cb, COMPONENT_Cr) sets the value of the adaptive primary conversion flag Apt_flag luminance to [COMPONENT_Y].
[0103]
　As a modification of formula (9), Apt_flag deriving unit 151 refers to the residual data existence flag cbf_luma luminance, a color difference adaptive primary conversion flag apt_flag [compID] as shown in the following equation (10) it may be derived.
[0104]
　apt_flag[compID] = cbf_luma ? apt_flag[COMPONENT_Y] : 0
　・・・（１０）
[0105]
　That, Apt_flag deriving unit 151, if the residual data existence flag cbf_luma luminance is 1 (true), the adaptive-primary conversion flag Apt_flag chrominance [compID] (compID = COMPONENT_Cb, COMPONENT_Cr) to the luminance of the adaptive primary conversion flag Apt_flag [ set the value of COMPONENT_Y], (when the luminance residual data existence flag cbf_luma of 0 (false)) other cases, set to 0.
[0106]
　In the case where apt_flag [COMPONENT_Y] does not exist on the encoded data, if the value of the flag is set to 0, it is not necessary to refer to the residual data existence flag of the luminance. However, Apt_flag if there is no operation for the initialization to 0 [COMPONENT_Y] is, Apt_flag the value of [COMPONENT_Y] becomes unstable, with reference to the residual data existence flag cbf_luma luminance, the luminance of the adaptive primary conversion flag Apt_flag it is desirable to control whether or not to refer to the value of [COMPONENT_Y].
[0107]
　Further, Pt_idx deriving unit 152 performs processing relating to the derivation of the primary conversion identifier Pt_idx color difference. For example, Pt_idx deriving unit 152, a color difference conversion block on adaptation primary conversion flag apt_flag [compID] (compID = COMPONET_Cb or COMPONENT_Cr), and on the basis of the primary conversion identifier pt_idx [COMPONENT_Y] relating to the luminance conversion block corresponding to the color difference conversion block the primary transform identifier pt_idx [compID] for the conversion block of the color difference, derived as the following equation (11).
[0108]
　pt_idx[compID] = apt_flag[compID] ? pt_idx[COMPONENT_Y] : 所定値
　・・・（１１）
[0109]
　That, Pt_idx deriving unit 152, when the color difference conversion block adaptive primary conversion flag relating apt_flag [compID] is 1 (true), and further, the primary conversion identifier pt_idx [compID] regarding the color difference conversion block, on the color difference conversion block set to the value of the corresponding primary transform identifier relating to the luminance conversion block pt_idx [COMPONENT_Y]. Otherwise, it sets a predetermined value.
[0110]
　As described above, the color difference adaptive primary conversion information deriving unit 131, an adaptive primary conversion information relating to a color difference conversion block may be derived based on adaptive primary conversion information relating to the luminance corresponding to the color difference conversion block. Thus, the adaptive primary conversion information relating to a color difference conversion block, it is not necessary to decode from the encoding data, it is possible to reduce the processing amount of decoding.
[0111]
　　
　Referring back to FIG. 10, the inverse primary transform selection portion 132, the prediction mode information pinfo, color signals identifier compID, adaptive primary conversion of the color signals corresponding to the color signal identifier compID flag apt_flag [compID], the color signal and enter the primary conversion identifier pt_idx color signals corresponding to the identifier compID [compID]. Inverse primary transform selection portion 132, the prediction mode information pinfo, color signals identifier compID, color signal adaptive primary conversion flag apt_flag identifier compID color signal corresponding to [compID], the primary conversion identifier of a color signal corresponding to the color signal identifier compID Referring to pt_idx [compID], then derive a transformation type TrTypeIdxH, and conversion type TrTypeIdxV inverse primary vertical conversion of the inverse primary horizontal conversion of the color signal specified by the color signal identifier compID, respectively, the inverse primary horizontal conversion unit 134 , and supplies the inverse primary vertical conversion unit 133.
[0112]
　Inverse primary vertical conversion unit 133, for each transform block of each color signal, transform coefficients Coeff_IS after the inverse secondary transform, inverse primary vertical conversion of the conversion type TrTypeIdxV, log2TBHSize indicating the vertical width of the conversion block (logarithmic value of longitudinal width) and input. Inverse primary vertical conversion unit 133, the converted coefficient Coeff_IS after the inverse secondary transform may perform the inverse primary vertical conversion IPver determined by the vertical width of the conversion type TrTypeIdxV and conversion block, the conversion after the inverse primary vertical conversion results supplied to the inverse primary horizontal conversion unit 134 as a coefficient Coeff_IPver. Note that the inverse primary vertical conversion IPver is an inverse transformation of the primary vertical conversion PVER.
[0113]
　Inverse primary horizontal conversion unit 134, an input for each conversion block of each color signal, transform coefficients Coeff_IPver after the inverse primary vertical conversion, inverse primary horizontal conversion of the conversion type TrTypeIdxH, log2TBWSize indicating the width of the transform blocks (logarithmic value of the horizontal width) to. Inverse primary horizontal conversion unit 134, the converted coefficient Coeff_IPver after the inverse primary vertical conversion, performs an inverse primary horizontal transform IPhor determined by the width of the conversion type TrTypeIdxH a conversion block, reverse the result as the prediction residual D ' and outputs to the outside of the conversion unit 113 (for supplying to the arithmetic unit 114). Note that the inverse primary horizontal transform IPhor is an inverse transformation of the primary horizontal conversion Phor.
[0114]
　　
　Next, the color signal specified by the color signal identifier compID in the inverse primary transform selection portion 132 opposite the primary horizontal transform of transform type TrTypeIdxH, and the inverse primary vertical conversion type TrTypeIdxV derivation of It will be described in detail.
[0115]
　Specifically, when the color signal identifier compID adaptive primary conversion flag color signal specified by apt_flag [compID] is 1 (true), the inverse primary transform selection unit 132 refers to the prediction mode information pinfo, FIG four transformation set TrSet conversion set identifier TrSetSetIdx = 0 ... 3 shown in the first table, for each horizontal and vertical direction, selects the transformation set TrSet comprising orthogonal transformation to be a candidate for the primary conversion. Incidentally, the correspondence of values ​​of the corresponding identifiers TrTypeIdx the type of orthogonal transformation may be varied in extent practicable not limited to FIG.
[0116]
　Incidentally, adaptive if the primary conversion flag apt_flag [compID] is 0 (false), the inverse primary transform selection portion 132, to indicate that no running adaptive primary conversion, inverse primary horizontal transform of transform type TrTypeIdxH, and inverse primary vertical the conversion of the conversion type TrTypeIdxV, predetermined (e.g., DCT-II) to set the value of the conversion type identifier indicating the type of orthogonal transform (TrIdxTypeH = TrIdxTypeV = predetermined value).
[0117]
　　
　if prediction type CuPredMode of CU conversion processing target block belongs indicates intra prediction (CuPredMode == MODE_INTRA), as shown in Table (LUT_IntraModeToTrSet) shown in FIG. 3, the intra prediction mode (IntraPredMode) It is performed on the basis of the. For example, implemented to set the converted set identifier TrSetIdx specify the following as shown in equation (12) and (13), transformation set TrSet that for each direction of transformation set (TrSetH, and TrSetV), the corresponding It is. Also, if the prediction type CuPredMode of CU conversion processing target block belongs is the inter prediction, as shown in the following expression (14), conversion set identifier TrSetIdx specifying the transformation set TrSet for inter prediction (= InterTrSetIdx) is set It is. Incidentally, the intra prediction mode IntraPredMode, if luminance, are set to the value of the luminance intra prediction mode IntraPredModeY, if the color difference, it is assumed that the value of the chrominance intra-prediction mode IntraPredModeC is set.
[0118]
　if (CuPredMode == MODE_INTRA ){
　　TrSetH = LUT_IntraModeToTrSet [ IntraPredMode ] [ H(=0) ]　・・・（１２）
　　TrSetV = LUT_IntraModeToTrSet [ IntraPredMode ] [ V(=1) ]　・・・（１３）
　} else { // CuPredMode==MODE_INTER
　　TrSetH = InterTrSetIdx
　　TrSetV = InterTrSetIdx　・・・（１４）
　}
[0119]
　Here, TrSetH represents the transformation set of primary horizontal conversion PThor (referred to as the primary horizontal transformation set), TrSetV shows transformation set of primary vertical conversion PTver (also referred to as a primary vertical conversion sets). Also, the look-up table LUT_IntraModeToTrSet is a correspondence table of FIG. Lookup table LUT_IntraModeToTrSet [] [] 1 th sequence of the intra prediction mode IntraPredMode as arguments, the second array, and argument {H = 0, V = 1}.
[0120]
　For example, in the case of the intra prediction mode number 18 (IntraPredMode == 18), as the primary horizontal transformation set TrSetH, transformation set of transformation set identifier TrSetIdx = 2 shown in Table (LUT_TrSetToTrTypeIdx) of FIG. 1 is selected, the primary vertical conversion set as TrSetV, transformation set of transformation set identifier TrSetIdx = 0 shown in the table of FIG. 1 is selected.
[0121]
　In FIG. 3, the intra prediction mode IntraPredMode found intra block copy; may exhibit (IBC Intra Block Copy, IntraBC, also referred to as a screen motion compensation) (on the table of FIG. 3, IntraPredMode == 67 is true IntraBC ), may be converted set identifier inter prediction only is assigned. Generally, the inter-prediction and intra-block copy, since the trend of the residual are similar, it is reasonable to assign the same transformation set. It is also possible to convert a set identifier of the intra block copy only is assigned.
[0122]
　　
　Furthermore, the inverse primary transform selection portion 132, the primary conversion identifier pt_idx color signal specified by the color signal identifier compID [compID] (compID = COMPONENT_Y , COMPONENT_Cb, COMPONENT_Cr) from, for example, according to the following equation (15) to derive a primary horizontal conversion designation flag pt_hor_flag and primary vertical conversion designation flag Pt_ver_flag.
[0123]
　pt_hor_flag = pt_dx[compID] & 0x01
　pt_ver_flag = pt_idx[compID] >> 1 (= (pt_idx[compID] & 0x10)>>1)
　・・・（１５）
[0124]
　That is, the primary conversion identifier pt_idx [compID] takes 2-bit value, the higher 1 bit is equivalent to the primary vertical conversion designation flag Pt_ver_flag, lower 1 bit is equivalent to the primary horizontal conversion designation flag Pt_hor_flag.
[0125]
　　
　Furthermore, the inverse primary transform selection portion 132, transformation set TrSetH selected for each horizontal / vertical directions, from TrSetV, orthogonal transform used for the inverse primary transform the conversion type, each are selected by the primary horizontal conversion designation flag Pt_hor_flag, and primary vertical conversion designation flag Pt_ver_flag.
[0126]
　Specifically, the inverse primary transform selection portion 132, a transform type TrTypeIdxH orthogonal transformation to be applied to the inverse primary horizontal transform, as shown in the following expression (16), the correspondence table of the conversion type as transformation set shown in FIG. 1 (LUT_TrSetToTrTypeIdx), determined on the basis of the primary horizontal transformation set TrSetH, and the primary horizontal conversion designation flag Pt_hor_flag.
[0127]
　TrTypeIdxH = LUT_TrSetToTrTypeIdx [ TrSetH ] [ pt_hor_flag ]
　・・・（１６）
[0128]
　Similarly, the inverse primary transform selection portion 132, a transform type TrTypeIdxV orthogonal transformation to be applied to the inverse primary vertical conversion, as shown in the following equation (17), the correspondence table of the transformation set the conversion type shown in FIG. 1 (LUT_TrSetToTrTypeIdx ), determined on the basis of the primary vertical transformation set TrSetVx, and the primary vertical conversion designation flag Pt_ver_flag.
[0129]
　TrTypeIdxV = LUT_TrSetToTrTypeIdx [ TrSetV ] [ pt_ver_flag ]
　・・・（１７）
[0130]
　For example, if the value of the transformation set identifier TrSetIdx indicated by the primary horizontal transformation set TrSetH 2, orthogonal to apply to the primary horizontal conversion from the conversion set a conversion set identifier TrSetIdx == 2 on transformation set definition table LUT_TrSetToTrTypeIdx in FIG conversion is selected. That is, if the primary horizontal conversion designation flag pt_hor_flag is 0, the horizontal conversion type identifier TrTypeIdxH that specifies the type of orthogonal transform of the primary horizontal conversion PThor, as shown in Table 1, the conversion type identifier TrTypeIdx showing a DST-VII is set to the value "4", if the primary horizontal conversion designation flag pt_hor_flag is 1, the horizontal conversion type identifier TrTypeIdxH, the value of the conversion type identifier TrTypeIdx showing a DCT-V "1" is set.
[0131]
　The conversion set identifier TrSetIdx of the table shown in FIG. 1, the primary horizontal conversion designation flag, and conversion type determined by the value of the primary vertical conversion specifying flag, it is allowed to be freely changed in extent practicable good.
[0132]
　Then, the inverse primary transform selection portion 132, converts the type TrTypeIdxH inverse primary horizontal transform IPThor color signal specified by the color signal identifier compID, and conversion type TrTypeIdxV inverse primary vertical conversion IPTver, respectively, the inverse primary vertical conversion unit 133 and supplied to the inverse primary horizontal conversion unit 134.
[0133]
　As described above, the inverse primary transform unit 123, in a case where the residual signal of the color difference is similar residual signal and trends of the luminance, the inverse adaptive primary transform selected in the luminance, applying the transform block of the color difference it is possible. Therefore, compared with the prior art, with respect to the residual signal of the color difference, improved inverse primary transform coding processing efficiency can be performed.
[0134]
　The luminance / adaptive primary conversion for each color difference flag Apt_flag, and as compared with a case where explicitly decode the primary conversion identifier Pt_idx, while suppressing a decrease in coding efficiency for a color difference, it is possible to reduce the processing amount of the encoder it can.
[0135]
　　
　Next, a flow of each process executed by the image decoding apparatus 100 will be described. First, with reference to the flowchart of FIG. 12, an example of a flow of the image decoding processing.
[0136]
　When the image decoding process is started, in step S101, the decoding unit 111 decodes the bit stream supplied to the image decoding apparatus 100 (coded data), the header information Hinfo, prediction mode information pinfo, conversion information TINFO, residual information rinfo, obtain information such as quantized transform coefficient levels level.
[0137]
　In step S102, the inverse quantization unit 112, and the transform coefficients Coeff_IQ inversely quantizes the quantized transform coefficient levels level obtained by the processing in step S101. The inverse quantization is the inverse process of the quantization performed in the image encoding process to be described later is the same processing as the inverse quantization performed in the picture coding processing.
[0138]
　In step S103, the inverse transform unit 113 inversely transforms the transform coefficients Coeff_IQ obtained by the processing in step S102, to derive a prediction residual D '. This inverse transform is the inverse process of the conversion process performed in the image encoding process to be described later, the inverse transform and the same processing performed in the picture coding processing.
[0139]
　In step S104, the prediction unit 116, based on the prediction mode information pinfo, making predictions in the prediction and the same prediction mode for encoding, to the predicted image generation.
[0140]
　In step S105, the arithmetic unit 114 processes the prediction residual D 'obtained by the step S103, it adds the predicted image obtained by the processing in step S104, obtaining a decoded image.
[0141]
　In step S106, the calculation unit 114 outputs the decoded image obtained by the processing in step S105 to the outside of the image decoding apparatus 100.
[0142]
　In step S107, the frame memory 115 stores the decoded image obtained by the processing in step S105.
[0143]
　When the process of step S107 is completed, the image decoding process is terminated.
[0144]
　　
　In step S101 of FIG. 12, as described above, various information included in the coded data # 1 is decoded. At that time, the decoding unit 111 performs adaptive primary conversion flag apt_flag and primary transform identifier information also decoding appropriate such Pt_idx. For example, the decoding unit 111 decodes the luminance adaptive primary conversion flag apt_flag [COMPONENT_Y] and brightness primary transform identifier pt_idx of [COMPONENT_Y]. However, as described above, in the coded data # 1, the adaptive primary conversion flag apt_flag chrominance [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and primary transform identifier pt_idx chrominance [compID] (compID = COMPONET_Cb or COMPONENT_Cr) It is not included. Accordingly, the decoding unit 111 decodes the pieces of information will be omitted.
[0145]
　In order to perform such decoding, the decoding unit 111, in step S101, it executes the primary conversion information decoding process. With reference to the flowchart of FIG. 13, an example of the flow of the primary conversion information decoding process.
[0146]
　When the primary conversion information decoding process is started, the decoding unit 111, in step S111, the processing object component is a luminance determines (compID == COMPONENT_Y) or. If it is determined that the luminance, the process proceeds to step S112.
[0147]
　In step S112, the decoding unit 111 may decode the encoded data # luminance adaptive primary conversion flag apt_flag contained 1 [COMPONENT_Y]. In step S113, the decoding unit 111 may decode the encoded data # Primary transformation identifier pt_idx luminance contained in 1 [COMPONENT_Y]. When the process of step S113 is completed, the primary conversion information decoding processing ends.
[0148]
　Further, in step S111, if the processing target component is determined to be color difference (compID! = COMPONENT_Y), the processing of step S112 and step S113 is omitted, the primary conversion information decoding processing ends. That is, in the case of the color difference, the decoding of the adaptive primary conversion flag apt_flag and primary transform identifier pt_idx is omitted. Therefore, it is possible to suppress an increase in the decoding load.
[0149]
　　
　Next, an example of the inverse conversion processing flow executed in step S103 of FIG. 12 will be described with reference to the flowchart of FIG. 14. When inverse transform processing is started, the switch 121 in step S121, whether the conversion the skip flag ts_flag is 2D_TS (2-dimensional transform skip mode), or, is transformed and quantized bypass flag transquant_bypass_flag is 1 (true) determines whether or not. Is converted skip identifier ts_idx is 2D_TS, or, if the conversion quantization bypass flag is determined to be 1 (true), the output transform coefficient Coeff_IQ by a switch 121 to the outside as a prediction residual D '(arithmetic unit 114 is supplied), the inverse transform process is completed, the process returns to FIG. 12.
[0150]
　Further, in step S121, converts the skip identifier ts_idx is not 2D_TS (mode other than the two-dimensional transform skip), and, when the transformed and quantized bypass flag is determined to be 0 (false), the conversion coefficient Coeff_IQ by the switch 121 is supplied to the inverse secondary transform unit 122, processing proceeds to step S122.
[0151]
　In step S122, the inverse secondary transform unit 122 subjects the input transform coefficients Coeff_IQ, based on the secondary conversion identifier St_idx, it performs an inverse secondary transform to derive a conversion factor Coeff_IS after the inverse secondary transform to output.
[0152]
　In step S123, the inverse primary transform selection portion 132, the color signal identifier compID is either a luminance, it is determined whether the color differences. When the color signal identifier compID shows the color difference (compID! = COMPONENT_Y), the processing proceeds to step S124. In step S124, the color difference adaptive primary conversion information deriving unit 131, adaptive primary conversion flag related to the color difference conversion block apt_flag [compID] (compID = COMPONET_Cb or COMPONENT_Cr), and primary converted identifier pt_idx relating to the luminance conversion block corresponding to the color difference conversion block based on the [COMPONENT_Y], to derive a primary transform identifier pt_idx [compID] for the conversion block of the color difference. Processing as the steps S124 ends, the process proceeds to step S125.
[0153]
　Further, in step S123, when the color signal identifier compID represents luminance (compID == COMPONENT_Y) the process proceeds to step S125.
[0154]
　In step S125, the inverse primary transform selection portion 132, the prediction mode information pinfo, color signals identifier compID, color signal adaptive primary conversion flag apt_flag identifier compID color signal corresponding to [compID], the color signal corresponding to the color signal identifier compID with reference to the primary conversion identifier pt_idx [compID], to derive a transform type TrTypeIdxH, and conversion type TrTypeIdxV inverse primary vertical conversion of the inverse primary horizontal conversion of the color signal specified by the color signal identifier compID.
[0155]
　In step S126, the inverse primary vertical conversion unit 133, for each transform block designated by the color signal identifier compID, the converted coefficient Coeff_IS after the inverse secondary transform of the transform block, a transform type TrTypeIdxV inverse primary vertical conversion performing an inverse primary vertical conversion IPver determined by the vertical width of the conversion block, and outputs a transform coefficient Coeff_IPver after the inverse primary vertical conversion results.
[0156]
　In step S127, the inverse primary horizontal conversion unit 134, for each transform block designated by the color signal identifier compID, the converted coefficient Coeff_IPver after the inverse primary vertical conversion of the conversion block, transformation types inverse primary horizontal transform TrTypeIdxH and performing an inverse primary horizontal transform IPhor determined by the height of the conversion block, and outputs the result as the prediction residual D '. When the process of step S127 is completed, the inverse transform process is completed, the process returns to FIG. 12.
[0157]
　　
　Next, an example of the flow chrominance adaptive primary conversion information deriving process executed in step S124 of FIG. 14 will be described with reference to the flowchart of FIG. 15. When the color difference adaptive primary conversion information deriving process is started, Apt_flag derivation unit 151 of the color difference adaptive primary conversion information deriving unit 131, in step S131, the the value of the adaptive primary conversion flag apt_flag [compID] chrominance, luminance adaptive primary It sets the value of the conversion flag apt_flag [COMPONENT_Y].
[0158]
　In step S132, Pt_idx deriving unit 152 of the color difference adaptive primary conversion information deriving unit 131, adaptive primary conversion flag apt_flag of the color difference [compID] determines whether it is true. Processing is determined to be true, the process proceeds to step S133. In step S133, Pt_idx deriving unit 152, the primary conversion identifier Pt_idx chrominance [compID], it sets the value of the primary conversion identifier Pt_idx luminance [COMPONENT_Y]. When the process of step S133 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0159]
　Further, in step S132 of FIG. 15, if the adaptive primary conversion flag apt_flag chrominance [compID] is determined to be false, the process proceeds to step S134. In step S134, Pt_idx deriving unit 152, the primary conversion identifier Pt_idx chrominance [compID], it sets a predetermined value. When the process of step S134 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0160]
　　
　Next, an example of the inverse primary transform selection the flow of the processing executed in step S125 of FIG. 14 will be described with reference to the flowchart of FIG. 16.
[0161]
　Conversely primary transform selection process is started, the inverse primary transform selection portion 132, in step S141, whether the adaptive primary conversion flag apt_flag color signals corresponding to the color signal identifier compID [compID] is 1 (true) or not determines whether or not. If the adaptive primary conversion flag apt_flag [compID] is determined to be 1 (true), the process proceeds to step S142.
[0162]
　In step S142, the inverse primary transform selection portion 132, based on the prediction mode information pinfo, inverse primary vertical conversion transformation set TrSetV (primary horizontal conversion sets), and the inverse primary horizontal transform transformation set TrSetH (primary vertical conversion sets) of to select.
[0163]
　In step S143, the inverse primary transform selection unit 132 derives from the primary conversion identifier pt_idx [compID] of the corresponding color signal to a color signal identifier compID, primary horizontal conversion designation flag Pt_hor_flag, and the primary vertical conversion designation flag Pt_ver_flag.
[0164]
　In step S144, the inverse primary transform selection unit 132 refers to the primary horizontal transformation set TrSetH and primary horizontal conversion designation flag Pt_hor_flag, selects a transform type TrTypeIdxH orthogonal transformation to be applied as an inverse primary horizontal transform IPThor.
[0165]
　In step S145, the inverse primary transform selection unit 132 refers primary vertical transformation set TrSetV, and the primary vertical conversion designation flag Pt_ver_flag, selects a transform type TrTypeIdxV orthogonal transformation to be applied as an inverse primary vertical conversion IPTver. When the process of step S145 is completed, the inverse primary transform selection process is completed, the process returns to FIG. 14.
[0166]
　Further, in step S141, if the adaptive primary conversion flag apt_flag [compID] is determined to be 0 (false), the process proceeds to step S146. In step S146, the inverse primary transform selection portion 132, as the conversion type TrTypeIdxH inverse primary horizontal transform IPThor, selects a predetermined orthogonal transformation (TrTypeIdxH = predetermined value).
[0167]
　In step S147, the inverse primary transform selection portion 132, as the conversion type TrTypeIdxV inverse primary vertical conversion IPTver, selects a predetermined orthogonal transformation (TrTypeIdxV = predetermined value). When the process of step S147 is completed, the inverse primary transform selection process is completed, the process returns to FIG. 14.
[0168]
　In other words, the inverse primary transform selection portion 132, in a manner corresponding to the value of the adaptive primary conversion flag apt_flag [compID] chrominance, the transformation type TrTypeIdxH inverse primary horizontal transform IPThor, and conversion type TrTypeIdxV inverse primary vertical conversion IPTver derivation to.
[0169]
　As described above, the inverse transform unit 113 to the image decoding apparatus 100, in a case where the residual signal of the color difference is similar residual signal and trends of the luminance, the inverse adaptive primary transform selected in the luminance, conversion of the color difference it is possible to apply to the block. Therefore, compared with the prior art, with respect to the residual signal of the color difference, improved inverse primary transform coding processing efficiency can be performed. The reduction, luminance / adaptive primary conversion for each color difference flag Apt_flag, and as compared with a case where explicitly decode the primary conversion identifier Pt_idx, while suppressing a decrease in coding efficiency for a color difference, the throughput of the encoding or decoding can do.
[0170]
　　
　In addition, the prediction type of the derivation of the adaptive primary conversion flag of the color difference may be limited to the case of inter CU. For example, as shown in the following expression (18), adaptive-primary conversion flag apt_flag regarding the color difference conversion block to be processed [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and adaptive relating to the luminance conversion block corresponding to the transform block of the color difference Primary conversion flag apt_flag [COMPONENT_Y], and may be derived based on the prediction type CuPredMode of the color difference conversion block belongs CU (coding unit).
[0171]
　if ( CuPredMode == MODE_INTER ) {
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else {
　 apt_flag[compID] = 0
　}
　・・・（１８）
[0172]
　For example, if prediction type CuPredMode of CU Field of the color difference conversion block is an inter prediction, (CuPredMode == MODE_INTER), adaptive primary conversion flag apt_flag regarding the color difference conversion block [compID] is, brightness corresponding to the transform block of the color difference It is set based on the adaptive primary conversion flag apt_flag about transformation block [COMPONENT_Y]. In contrast, when the prediction type CuPredMode of CU Field of the color difference conversion block is not inter prediction (an intra prediction), the value of the adaptive primary conversion flag apt_flag regarding the color difference conversion block [compID] is set to 0 (false) that.
[0173]
　An example of the flow of the color difference adaptive primary conversion information deriving process in such a case will be described with reference to a flowchart of FIG. 17. This flowchart corresponds to the flowchart of FIG. 15. In this case, the color difference adaptive primary conversion information deriving process is started, Apt_flag derivation unit 151, at step S151, the prediction type CUPredMode of CU conversion processing target block belongs whether the inter prediction (MODE_INTER) judge. If it is determined that the inter prediction, the processing proceeds to step S152.
[0174]
　In step S152, Apt_flag derivation unit 151, as in step S131 of FIG. 15, the value of the adaptive primary conversion flag apt_flag [compID] chrominance, sets the value of the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] . Processing and the processing of step S152 is completed, the process proceeds to step S154.
[0175]
　Further, in step S151, the case where prediction type CUPredMode of CU conversion processing target block belongs is determined to be the intra prediction, the process the process proceeds to the step S153. In step S153, Apt_flag deriving unit 151, the value of the adaptive primary conversion flag apt_flag [compID] color difference to 0 (false). Processing and the processing of step S153 is completed, the process proceeds to step S154.
[0176]
　Each processing of steps S154 through step S156 are performed in the same manner as the processes of steps S132 to step S134 of FIG. 15. When the process in step S155 or step S156 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0177]
　If the prediction type is intra prediction, the luminance and color difference, there is a case where the intra prediction mode is different. At this time, since the tendency of the luminance and chrominance of the residual signal it is different, it is possible to lower the coding efficiency when applying adaptive primary conversion information of the luminance to chrominance. Therefore, run the color difference adaptive primary conversion information deriving process as in FIG. 17, if prediction type is the inter prediction only, by so as to apply an adaptive primary conversion information of the luminance to the color difference, compared with the case of FIG. 15 it can be further suppressed reduction of coding efficiency Te.
[0178]
　　
　In addition, the derivation of the adaptive primary conversion flag of the color difference, the inter CU, or intra CU and luminance intra prediction mode may be limited when the intra block copy (screen motion compensation) . For example, as shown in the following expression (19), adaptive-primary conversion flag apt_flag regarding the color difference conversion block to be processed [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and adaptive relating to the luminance conversion block corresponding to the transform block of the color difference Primary conversion flag apt_flag [COMPONENT_Y], and the prediction type CuPredMode of the color difference conversion block belongs CU (coding unit), and may be set based on the prediction mode information pinfo.
[0179]
　if ( CuPredMode == MODE_INTER ) {
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else if (IntraPredModeC == "IntraBC") { // CuPredMODE == MODE_INTRA
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else {
　 apt_flag[compID] = 0
　}
　・・・（１９）
[0180]
　That is, if prediction type CuPredMode of CU Field of the color difference conversion block is an inter prediction or prediction type of CU is intra prediction, and, when intra prediction mode IntraPredModeC color difference is an intra block copy (IntraBC) the color difference conversion block on adaptation primary conversion flag apt_flag [compID] is set based on the adaptive primary conversion flag apt_flag [COMPONENT_Y] for the corresponding luminance conversion block into transform blocks of the color difference. In contrast, otherwise (a prediction type CuPredMode the CU is intra prediction, and, when intra prediction mode IntraPredModeC color difference is not an intra block copy), adaptive-primary conversion regarding the color difference conversion block flag apt_flag of [compID] the value is set to 0 (false). In addition, if it contains luminance and chrominance information in the same CU, in the above equation (19), in place of the intra prediction mode IntraPredModeC color difference may be a condition determined using the intra prediction mode IntraPredModeY luminance .
[0181]
　An example of the flow of the color difference adaptive primary conversion information deriving process in such a case will be described with reference to the flowchart of FIG. 18. This flowchart corresponds to the flowchart of FIG. 15. In this case, the color difference adaptive primary conversion information deriving process is started, Apt_flag deriving unit 151, in step S171, the prediction type CUPredMode of CU conversion processing target block belongs whether the inter prediction (MODE_INTER) judge. If it is determined that the intra prediction, the processing proceeds to step S172.
[0182]
　In step S172, Apt_flag deriving unit 151, an intra prediction mode IntraPredModeC chrominance determines whether an intra block copy (IntraBC). If it is determined that the intra-block copy (IntraBC), the process proceeds to step S173. Further, in step S171, if the prediction type CUPredMode of CU conversion processing target block belongs is determined to be inter prediction (MODE_INTER), the process proceeds to step S173.
[0183]
　In step S173, Apt_flag derivation unit 151, as in step S131 of FIG. 15, the value of the adaptive primary conversion flag apt_flag [compID] chrominance, sets the value of the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] . Processing and the processing of step S173 is completed, the process proceeds to step S175.
[0184]
　Further, in step S172, if the intra prediction mode IntraPredModeC the color difference is determined not intra block copy (IntraBC), the process the control process goes to step S174. In step S174, Apt_flag deriving unit 151, the value of the adaptive primary conversion flag apt_flag [compID] color difference to 0 (false). Processing and the processing of step S174 is completed, the process proceeds to step S175.
[0185]
　Each processing of steps S175 through step S177 are performed in the same manner as the processes of steps S132 to step S134 of FIG. 15. When the process in step S176 or step S177 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0186]
　The prediction type is a intra prediction, if the intra prediction mode IntraPredMode is intra block copy, similarly to the inter prediction, trend of the residual and residual chrominance luminance similar. Therefore, as compared with , and the prediction type is a intra prediction, if the intra prediction mode IntraPredMode is IntraBC, it is possible to adapt the primary conversion of the luminance is applied to the color difference, Furthermore it possible to improve the coding efficiency.
[0187]
　　
　In addition, the derivation of the adaptive primary conversion flag of the color difference, the inter CU or, may be limited when the intra CU and luminance intra prediction mode is equal to the chrominance intra-prediction mode. For example, as shown in the following expression (20), adaptive-primary conversion flag apt_flag regarding the color difference conversion block to be processed [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and adaptive relating to the luminance conversion block corresponding to the transform block of the color difference Primary conversion flag apt_flag [COMPONENT_Y], even as prediction type CuPredMode, and luminance intra prediction mode IntraPredModeY of the CU of CU that the color difference conversion block belongs (coding unit) is set based on the color difference intra mode IntraPredModeC of the CU good.
[0188]
　if ( CuPredMode == MODE_INTER ) {
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else if (IntraPredModeY == IntraPredModeC) { // CuPredMode==MODE_INTRA
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else {
　 apt_flag[compID] = 0
　}
　・・・（２０）
[0189]
　In other words, the prediction type CuPredMode of CU Field of the color difference conversion block inter prediction, or prediction type of CU intra prediction, and, when intra prediction mode IntraPredModeC color difference is equal to the intra prediction mode IntraPredModeY luminance for the color difference conversion block adaptive primary conversion flag apt_flag [compID] is set based on the adaptive primary conversion flag apt_flag [COMPONENT_Y] relating to the luminance conversion block corresponding to the transform block of the color difference. In contrast, otherwise (prediction type CuPredMode intra prediction of CU, and, if the intra prediction mode of the color difference is different from the intra-prediction mode of the luminance), adaptive-primary conversion regarding the color difference conversion block flag apt_flag of [compID] the value is set to 0 (false).
[0190]
　An example of the flow of the color difference adaptive primary conversion information deriving process in such a case will be described with reference to the flowchart of FIG. 19. This flowchart corresponds to the flowchart of FIG. 15. In this case, the color difference adaptive primary conversion information deriving process is started, Apt_flag deriving unit 151, in step S181, the prediction type CUPredMode of CU conversion processing target block belongs whether the inter prediction (MODE_INTER) judge. If it is determined that the intra prediction, the processing proceeds to step S182.
[0191]
　In step S182, Apt_flag derivation unit 151 determines the intra prediction mode IntraPredModeC color difference whether equal intra prediction mode IntraPredModeY brightness. If it is determined to be equal, the process proceeds to step S183. Further, in step S182, if the intra prediction mode IntraPredModeC color difference is determined to not equal to the intra prediction mode IntraPredModeY luminance, the process proceeds to step S184.
[0192]
　Each processing of steps S183 through step S187 are performed in the same manner as the processes of step S173 to step S177 in FIG. 18. When the process in step S186 or step S187 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0193]
　If the prediction type of CU is intra prediction, the luminance and color difference, when the intra prediction mode are equal, the tendency of the luminance and chrominance of the residual signal are similar. Therefore, only when the intra prediction mode and the chrominance intra-prediction mode of the luminance are equivalent, the adaptive primary conversion information of the luminance, by applying to the color difference, as compared with the case of the , prediction type there a intra prediction, if the intra prediction mode and the chrominance intra-prediction mode of the luminance are equivalent, the adaptive primary conversion of the luminance, it is possible to apply to the color difference, further improve the coding efficiency be able to.
[0194]
　　
　Furthermore, the information on adaptation primary conversion of the color difference, adds a flag indicating whether or not to set based on the adaptive primary conversion of luminance, the setting information on adaptation primary conversion of the color difference by the flag it may be controlled. For example, the following as shown in equation (21), adaptive-primary conversion regarding the color difference conversion block to be processed flag apt_flag [compID] a (compID = COMPONET_Cb or COMPONENT_Cr), adaptive-primary conversion of the luminance transform block corresponding to the color difference conversion block so as to set based on the color difference adaptive primary conversion information estimation flag chroma_apt_infer_flag indicating whether to set (estimated) from the information, luminance conversion adaptive for block-primary conversion flag apt_flag corresponding to transform blocks of the color difference [COMPONENT_Y] and it may be.
[0195]
　if ( chroma_apt_info_infer_flag [compID] ) {
　 apt_flag[compID] = apt_flag[COMPONENT_Y] : 0
　} else {
　 apt_flag[compID] = 0
　}
　・・・（２１）
[0196]
　That is, if the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID] is 1 (true), the color difference conversion block on adaptation primary conversion flag apt_flag [compID] is adaptive primary conversion relating to the luminance conversion block corresponding to the transform block of the color difference It is set based on the flag apt_flag [COMPONENT_Y]. In contrast, (if the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID] is 0 (false)) otherwise, the value of the adaptive primary conversion regarding the color difference conversion block flag apt_flag [compID] is 0 (false) It is set.
[0197]
　An example of a syntax of TU in this case A of FIG. 20. Semantics of each parameter is as shown in B in FIG. 20. Further, FIG. 21 shows an example of the syntax of residual_prediction_flag coding included in the syntax of A in FIG. 20. Color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag along the defined syntax table shown in FIG. 21, is decoded by the decoding unit 111.
[0198]
　That is, in the case the color signal identifier compID indicates the color difference, and a luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] is 1 (true), and converts the skip of the color signal corresponding to the color signal identifier compID a flag ts_flag [compID] is 0 (false), and, when the transformed and quantized bypass flag flag is 0 (false), the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag is decoded. For other conditions, the value of the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag is 0 estimated.
[0199]
　Decoding unit 111, in step S101 of FIG. 12, by executing a color difference adaptive primary conversion information estimation flag decoding process performs such decoding. With reference to the flowchart of FIG. 22 illustrating an example of a flow of the color difference adaptive primary conversion information estimation flag decoding.
[0200]
　Determining the color difference adaptive primary conversion information estimation flag decoding process is started, the decoding unit 111, in step S191, whether the color signal identifier compID is luminance. If the chrominance signal identifier compID is determined to be a color difference (compID! = COMPONENT_Y), the process proceeds to step S192. In step S192, the decoding unit 111, the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] determines whether it is 1 (true). If the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] is determined to be 1 (true), the process the process proceeds to the step S193.
[0201]
　In step S193, the decoding unit 111, the transformed and quantized bypass flag Transquant_bypass_flag, determines whether or not 1 (true). If the conversion quantization bypass flag is determined to be 0 (false) (! Transquant_bypass_flag), the process proceeds to step S194. In step S194, the decoding unit 111, converts the skip flag ts_flag color signals corresponding to the color signal identifier compID [compID] determines whether it is 1 (true). If the conversion the skip flag ts_flag [compID] is determined to be 0 (false) (! Ts_flag [compID]), the process proceeds to step S195.
[0202]
　In step S195, the decoding unit 111, a bit string of coded data # 1, decodes the chrominance signal identifier compID corresponding to the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID], is output as part of the conversion information TINFO. When the process of step S195 is completed, the color difference adaptive primary conversion information estimation flag decoding process is completed, the process returns to FIG. 12.
[0203]
　Further, in step S191, when the color signal identifier compID is determined to be the luminance (compID == COMPONENT_Y), the process proceeds to step S196. Further, in step S192 (! Apt_flag [COMPONENT_Y]), if the luminance of the adaptive primary conversion flag apt_flag [COMPONENT_Y] is determined to be 0 (false), the process proceeds to step S196. Further, in step S193, when the conversion quantization bypass flag is determined to be 1 (true) (transquant_bypass_flag), the process proceeds to step S196. Further, in step S194, if the conversion the skip flag ts_flag [compID] is determined to be 1 (true) (ts_flag [compID]), the process proceeds to step S196.
[0204]
　In step S196, the decoding unit 111, omitting the decoding of the color signal identifier compID corresponding color difference adaptive primary conversion information estimation flag to chroma_apt_info_infer_flag [compID], it sets the value of the flag to 0 (false) (chroma_apt_info_infer_flag [compID] = 0). When the process of step S196 is completed, the color difference adaptive primary conversion information estimation flag decoding process is completed, the process returns to FIG. 12.
[0205]
　Color difference adaptive primary conversion information deriving unit 131 thus decoded or using the set color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID], performs color difference adaptive primary conversion information deriving process. An example of the flow of the color difference adaptive primary conversion information deriving process in such a case will be described with reference to the flowchart of FIG. 23. This flowchart corresponds to the flowchart of FIG. 15. In this case, the color difference adaptive primary conversion information deriving process is started, Apt_flag derivation unit 151 determines, in step S201, the value of the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID] is whether is true. If it is determined to be true, the process proceeds to step S202. Further, in step S201, if the value of the color difference adaptive primary conversion information estimation flag chroma_apt_info_infer_flag [compID] is determined to be false, the process proceeds to step S203.
[0206]
　Each processing of steps S202 through step S206 are performed in the same manner as the processes of step S183 to step S187 of FIG. 19. When the process in step S205 or step S206 is completed, the color difference adaptive primary conversion information deriving process is completed, the process returns to FIG. 14.
[0207]
　Relative trend different signal residual and residual color difference of the luminance, whether to apply adaptive primary conversion information of the luminance for the color difference, based on the color difference adaptive primary conversion information estimation flag explicitly it is possible to control. Therefore, as compared with the case of FIG. 15, a more relative to efficiently color difference, it is possible to apply an adaptive primary conversion, it is possible to improve the coding efficiency.
[0208]
　　
　In addition, the derivation of the adaptive primary conversion flag of the color difference, the short side size of the conversion block of the color difference may be limited if it is higher than a predetermined threshold value. For example, as shown in the following expression (22), adaptive-primary conversion flag apt_flag regarding the color difference conversion block to be processed [compID] (compID = COMPONET_Cb or COMPONENT_Cr) and adaptive relating to the luminance conversion block corresponding to the transform block of the color difference Primary conversion flag apt_flag [COMPONENT_Y], and may be set based on the size of the short sides of the color difference conversion block.
[0209]
　if ( min(log2TBWSize, log2TBHSize) >= TH ) {
　 apt_flag[compID] = apt_flag[COMPONENT_Y]
　} else {
　 apt_flag[compID] = 0
　}
　・・・（２２）
[0210]
　In other words, if the size of the short sides of the color difference conversion block is not less than the threshold value (min (log2TBWSize, log2TBHSize)> = TH), adaptive primary conversion flag apt_flag regarding the color difference conversion block [compID] is, brightness corresponding to the transform block of the color difference It is set based on the adaptive primary conversion flag apt_flag about transformation block [COMPONENT_Y]. In contrast, (when the size of the short side of the color difference conversion block is less than the threshold) Otherwise, the value of the adaptive primary conversion regarding the color difference conversion block flag apt_flag [compID] is set to 0 (false).
[0211]
　An example of the flow of the color difference adaptive primary conversion information deriving process in such a case will be described with reference to the flowchart of FIG. 24. This flowchart corresponds to the flowchart of FIG. 15. In this case, the color difference adaptive primary conversion information deriving process is started, Apt_flag derivation unit 151, at step S211, the size of the short side of the transform block processed (min (log2TBWSize, log2TBHSize)) is greater than a predetermined threshold value TH It determines whether or not it is. Size of the short side of the transform block processed is equal to or greater than a predetermined threshold value (min (log2TBWSize, log2TBHSize)> = TH) and when it is determined, the process proceeds to step S212. Further, in step S211, the size of the short side of the transform block processed is less than a predetermined threshold value (min (log2TBWSize, log2TBHSize)
　The vertical width of the transform block of the color difference, may be able to select a transform type of inverse primary horizontal transform IPhor and inverse primary vertical conversion IPver the size of the width. For example, in the above, the inverse primary transform selection portion 132, the correspondence table of the transformation set the conversion type shown in FIG. 1 (LUT_TrSetToTrTypeIdx), primary {horizontal, vertical} transformation set Trset {H, V}, and primary {horizontal, vertical} conversion designation flag pt_ {hor, based on the ver} _flag, have been described to determine the type of orthogonal transformation to be applied to the inverse transform in each direction, but are not limited thereto. For example, in the case of the transform block of the color difference, with further reference to the size of the color difference conversion block may be determined transformation type in each direction.
[0215]
　For example, in step S144 of FIG. 16, as shown in the following expression (23) (here, logarithm log2TBWSize breadth) lateral width of the color difference conversion block case it is less than the threshold value TH, (inverse) transformation types primary horizontal conversion the TrTypeIdxH set to a predetermined conversion type (DCT-2), otherwise, the horizontal transformation set TrSetH, based on the primary horizontal conversion designation flag Pt_hor_flag, so as to set the conversion type TrTypeIdxH of (inverse) primary horizontal transform it may be. In the equation (23), specific examples of the threshold value TH is set, for example, 1 or 2 and. In the equation (23), instead of the logarithmic value log2TBWSize the width of the transform block may be replaced by the width TBW. In that case, the threshold TH 'is set to 1 << TH.
[0216]
　an if (compID == COMPONENT_Y) {
　　　　　TrTypeIdxH = LUT_TrSetToTrTypeIdx [TrSetH] [Pt_hor_flag]
　} the else an if (Log2TBWSize <= TH) {
　　　　　TrTypeIdxH = predetermined value (set value converted type indicating DCT-II)
　} the else {
　　　　　TrTypeIdxH = LUT_TrSetToTrTypeIdx [TrSetH] [Pt_hor_flag]
　}
　· · · (23)
[0217]
　Such an example of the flow of the primary horizontal conversion type deriving processing executed in step S144 of FIG. 16 will be described with reference to the flowchart of FIG. 25. When the primary horizontal conversion type deriving process is started, the inverse primary transform selection portion 132 determines in step S221, the processing object whether the conversion block of luminance. If it is determined that the conversion block of the color difference, the process proceeds to step S222. In step S222, the inverse primary transform selection portion 132, the width of a color difference conversion block (logarithmic value log2TBWSize breadth) is equal to or less than the threshold value TH. If the width of a color difference conversion block is determined to be larger than the threshold value (log2TBWSize> TH), the process proceeds to step S223. Further, in step S221, if the processing target is determined to be a transform block of the luminance, the process proceeds to step S223.
[0218]
　In step S223, the inverse primary transform selection portion 132, (inverse) transformation type TrTypeIdxH primary horizontal translation, the correspondence table of the conversion type as transformation set shown in FIG. 1 (LUT_TrSetToTrTypeIdx), primary horizontal transformation set TrSetH, and primary horizontal transform set based on the designation flag Pt_hor_flag. When the process of step S223 is completed, the primary horizontal conversion type deriving process is completed, the process returns to FIG. 16.
[0219]
　Further, in step S222, if the width of the color difference conversion block is determined to be equal to or less than the threshold (log2TBWSize <= TH), the process proceeds to step S224. In step S224, the inverse primary transform selection portion 132, a transform type TrTypeIdxH of (inverse) primary horizontal conversion is set to a predetermined value. When the process of step S224 is completed, the primary horizontal conversion type deriving process is completed, the process returns to FIG. 16.
[0220]
　The same applies to the vertical direction. For example, in step S145 of FIG. 16, as shown in the following expression (24) (here, logarithm log2TBHSize vertical width) height of the color difference conversion block may equal to or smaller than the threshold value TH, the (inverse) primary vertical conversion set the predetermined conversion type (DCT-2) as the conversion type TrTypeIdxV, otherwise, vertical conversion set identifier TrSetV, based on the primary vertical conversion designation flag Pt_ver_flag, setting the transform type TrTypeIdxV of (inverse) primary vertical conversion it may be. In the equation (24), specific examples of the threshold value TH is set, for example, 1 or 2 and. Further, in the equation (24), in place of the logarithm log2TBHSize the height of the transform block may be replaced by vertical width TBH. In that case, the threshold TH 'is set to 1 << TH.
[0221]
　an if (compID == COMPONENT_Y) {
　　　　　TrTypeIdxV = LUT_TrSetToTrTypeIdx [TrSetV] [Pt_ver_flag]
　} the else an if (Log2TBHSize <= TH) {
　　　　　TrTypeIdxV = predetermined value (set value converted type indicating DCT-II)
　} the else {
　　　　　TrTypeIdxV = LUT_TrSetToTrTypeIdx [TrSetV] [Pt_ver_flag]
　}
　· · · (24)
[0222]
　Such an example of the flow of the primary vertical conversion type deriving processing executed in step S145 of FIG. 16 will be described with reference to the flowchart of FIG. 26. When the primary vertical conversion type deriving process is started, the inverse primary transform selection portion 132 determines in step S231, the processing object whether the conversion block of luminance. If it is determined that the conversion block of the color difference, the process proceeds to step S232. In step S232, the inverse primary transform selection portion 132, the vertical width of the color difference conversion block (logarithmic value log2TBHSize vertical width) is equal to or less than the threshold value TH. If the vertical width of a color difference conversion block is determined to be larger than the threshold value (log2TBHSize> TH), the process proceeds to step S233. Further, in step S231, if the processing target is determined to be a transform block of the luminance, the process proceeds to step S233.
[0223]
　In step S233, the inverse primary transform selection portion 132, (inverse) transformation type TrTypeIdxV primary vertical conversion, the correspondence table of the conversion type as transformation set shown in FIG. 1 (LUT_TrSetToTrTypeIdx), primary vertical transformation set TrSetV, and primary vertical conversion set based on the designation flag Pt_ver_flag. When the process of step S233 is completed, the primary vertical conversion type deriving process is completed, the process returns to FIG. 16.
[0224]
　Further, in step S232, if the width of the color difference conversion block is determined to be equal to or less than the threshold (log2TBHSize <= TH), the process proceeds to step S234. In step S234, the inverse primary transform selection portion 132, a transform type TrTypeIdxV of (inverse) primary vertical conversion is set to a predetermined value. When the process of step S234 is completed, the primary vertical conversion type deriving process is completed, the process returns to FIG. 16.
[0225]
　If the width of the transform block of the color difference (vertical width) is smaller than a predetermined value (for example 4), as (inverse) primary horizontal (vertical) conversion, DST-7 / DST-1 / DCT-5 / DCT-8 and DCT- in a 2, the difference in the coding efficiency is small. Therefore, when the lateral width of the transform block of the color difference (vertical width) is smaller than a predetermined value (for example 4), by selecting as the (inverse) primary horizontal (vertical) conversion, a predetermined orthogonal transformation (DCT-2), a predetermined it is possible to reduce the circuit scale related to orthogonal transform utilized in the adaptive primary conversion of less than a value size.

The scope of the claims
[Requested item 1]
Using information on the inverse orthogonal transform of the color difference which is derived based on the information for the inverse orthogonal transform of the luminance, the inverse orthogonal transform unit for performing the inverse orthogonal transformation of the color difference
　image processing apparatus comprising a.
[Requested item 2]
　Information relating to the inverse orthogonal transformation, including adaptive primary conversion flag indicating whether to apply the adaptive inverse primary transform is used as adaptively selecting and inverse primary transform one from among a plurality of different inverse orthogonal transform
　claim 1 the image processing apparatus according to.
[Requested item 3]
　The value of the adaptive primary conversion flag of the color difference is set to the value of the adaptive primary conversion flag of the luminance
　image processing apparatus according to claim 2.
[Requested item 4]
　Information relating to the inverse orthogonal transformation, including the primary translation identifier indicating whether to apply any inverse primary transform to the inverse primary transform in the vertical direction and the horizontal direction
　image processing apparatus according to claim 3.
[Requested item 5]
　The value of the primary conversion identifier of the color difference, when the adaptive primary conversion flag of the color difference is true, is set to the value of the primary conversion identifier of luminance, if the adaptive primary conversion flag of the color difference is false, set to a predetermined value It is
　an image processing apparatus according to claim 4.
[Requested item 6]
　The inverse orthogonal transform unit, as the inverse orthogonal transform, an inverse primary transform
　image processing apparatus according to claim 5.
[Requested item 7]
　The inverse orthogonal transform unit, as the inverse orthogonal transformation is performed and the inverse primary horizontal transform is the inverse primary transform in the horizontal direction, the inverse primary vertical conversion which is the inverse primary transform in the vertical direction
　image according to claim 6 processing apparatus.
[Requested item 8]
　The value of the adaptive primary conversion flag of the color difference, when the prediction type of the coding block transform block to be processed belongs is an inter prediction is set to the value of the adaptive primary conversion flag of luminance
　according to claim 2 image processing apparatus.
[Requested item 9]
　The value of the adaptive primary conversion flag of color difference, or the prediction type of the coding block transform block to be processed belongs is an inter prediction, or, if the prediction mode is the intra prediction of an intra block copy, the adaptive luminance It is set to the value of the primary conversion flag
　image processing apparatus according to claim 2.
[Requested item 10]
　If the value of the adaptive primary conversion flag of color difference, or the prediction type of the coding block transform block to be processed belongs is an inter prediction or an intra prediction is prediction mode in the luminance and color difference matches, luminance is set to the value of the adaptive primary conversion flag
　image processing apparatus according to claim 2.
[Requested item 11]
　The value of the adaptive primary conversion flag of the color difference, when the adaptive primary conversion flag brightness said adaptive primary conversion flag chrominance adaptive primary conversion information estimation flag indicating estimated based on the color difference is true, the adaptive luminance It is set to the value of the primary conversion flag
　image processing apparatus according to claim 2.
[Requested item 12]
　The value of the adaptive primary conversion flag of the color difference, when the short side size of the processed transform blocks of the color difference is not smaller than a predetermined threshold value is set to the value of the adaptive primary conversion flag of luminance
　according to claim 2 image processing apparatus.
[Requested item 13]
　The inverse orthogonal transform unit, as the inverse orthogonal transform, is configured to perform the inverse primary horizontal transform is inverse primary transform in the horizontal direction, the inverse primary vertical conversion which is inverse primary transform in the vertical direction,
　the inverse primary horizontal conversion type of conversion, when the size of the horizontal width of the processed transform blocks of the color difference is greater than a predetermined threshold value, is set based on the horizontal transformation set and the primary horizontal conversion designation flag,
　convert the type of the inverse primary vertical transform , if the size of the vertical width of the processed transform blocks of the color difference is greater than a predetermined threshold value is set based on the vertical conversion set and the primary vertical conversion designation flag
　image processing apparatus according to claim 1.
[Requested item 14]
　Information relating to the inverse orthogonal transformation involves conversion skip flag indicating whether to skip the inverse orthogonal transform processing
　image processing apparatus according to claim 1.
[Requested item 15]
　The value of the conversion the skip flag of the color difference is set to the value of the conversion the skip flag luminance
　image processing apparatus according to claim 14.
[Requested item 16]
　Information relating to the inverse orthogonal transformation, including secondary conversion identifier indicating whether to apply any inverse secondary transform
　image processing apparatus according to claim 1.
[Requested item 17]
　The value of the secondary transform identifier of the color difference is set to the value of the secondary transform identifier luminance
　image processing apparatus according to claim 16.
[Requested item 18]
　Using information on the inverse orthogonal transform of the color difference which is derived based on the information about the inverse orthogonal transformation brightness, performs the inverse orthogonal transformation of the color difference
　image processing method.
[Requested item 19]
　Using information on the orthogonal transformation of the color difference which is derived based on the information on the orthogonal transformation brightness, the orthogonal transform unit for performing the orthogonal transform of the color difference
　image processing apparatus comprising a.
[Requested item 20]
　Using information on the orthogonal transformation of the color difference which is derived based on the information on the orthogonal transformation brightness, performs the orthogonal transform of the color difference
　image processing method.