JP2019075678A - Image encoder, image decoder, image coding program, and image decoding program - Google Patents

Abstract

To restrain image quality deterioration resulting from quantization distortion.SOLUTION: An image encoder (1) related to first feature includes a conversion part (104) performing orthogonal conversion for a conversion block, i.e., a part of the residual image of a prediction image and an input image, an activation determination part (109b) for determining whether or not the conversion skip (TS mode) for skipping the orthogonal conversion is activated with regard to a 4×4 conversion block having a 4×4 size, in units of sequence consisting of one or more than one frames, and a quantization section (105) performing quantization for each 4×4 conversion block corresponding to the sequence determined that the conversion skip was activated by the activation determination part, without using a first quantization matrix consisting of elements of different values, but using a second quantization matrix consisting of elements of equal values.SELECTED DRAWING: Figure 1

Description

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

  A predictive coding scheme is known as a coding scheme that compresses the amount of information with high efficiency. The predictive coding scheme predicts a predicted image at the current time from an image coded in the past, and encodes a residual image which is a residual between the predicted image and the input image. The encoding of the residual image transforms the residual image, quantizes the transform coefficient indicated in the transform domain (for example, the spatial frequency domain) in this transform processing, and then performs lossless encoding such as entropy encoding. Say Here, since the low frequency band (low band) in the spatial frequency is more sensitive than the high band in human visual characteristics, often, the lower the low band in quantization, the larger the quantization width is used in the high band. There is. The data formed by the quantization width for each frequency in such a block is called a quantization matrix. By applying a quantization matrix that uses a smaller quantization width for lower frequencies and a larger quantization width for higher frequencies, low-frequency components of the residual signal that are visually sensitive are emphasized, and it is subjective even if the amount of information is compressed. It is possible to suppress the deterioration of quality.

  There is HEVC (High Efficiency Video Coding) method (also referred to as “ISO / IEC 23008-2 HEVC, ITU-T Recommendation H. 265”) as a representative method of the predictive coding method (Non-Patent Document 1). In the method, it is determined whether to skip (omit) orthogonal transformation, which is a type of transformation processing, for each transformation block that is part of the residual image, and when it is determined to skip, the residual image is directly quantized Lossless coding processing is used, and whether or not to perform orthogonal transformation is determined by comparing the magnitudes of residuals and signal characteristics for each, skipping such orthogonal transformation to perform quantization and coding The processing mode to perform the conversion is called the transform skip (TS) mode. In general, in the case of a signal in which the high-frequency component of the image signal is an important element as the coding quality, the high-frequency component is obtained by orthogonal transformation and quantization when a quantization matrix considering visual characteristics is applied. Therefore, the TS mode may be selected in order to reduce the amount of information while suppressing the degradation of high frequency components, and when the TS mode is selected, a quantum according to the position coordinates in the block By applying the quantization width, even in the case of the signal characteristic in which the high frequency component of the residual image is the main, the information amount can be efficiently compressed without losing the quality.

  Here, in a general-purpose profile (a profile called Main such as Main profile or Main 10 pforile) shown in Non-Patent Document 1, an arbitrary quantization in encoding of 4 × 4 transform block having a block size of 4 × 4 is arbitrary. A quantization matrix (hereinafter referred to as “first quantization matrix” as appropriate) having a smaller quantization width in the lower band in addition to the quantization matrix, or a quantum having the same quantization width regardless of the coordinates or frequency in the block It is defined to use a quantization matrix (hereinafter referred to as “second quantization matrix” as appropriate).

Recommendation ITU-T H.2. 265, (04/2013) “, High efficiency video coding”, International Telecommunication Union, April 2013

  However, there is no indication as to which quantization matrix should be used depending on whether or not orthogonal transformation is performed for a 4 × 4 transform block, and whether it is in TS mode or not, either quantization matrix Are fixedly applied in units of frames or sequences (a plurality of series of frames).

  If the first quantization matrix, which is an arbitrary quantization matrix that is a quantization matrix having a smaller quantization width in the lower range, is used for quantization, visually reducing the amount of information when applying orthogonal transformation In the case where orthogonal transformation is skipped, the block to which the quantization matrix is applied is not orthogonally transformed, and therefore, spatial distortion (quantization distortion (quantization distortion may be caused by different quantization depending on coordinates). ), For example, there is a problem that visually noticeable deterioration occurs at the boundary with adjacent blocks. Such a phenomenon tends to occur in the case where the low-frequency component of the residual image is not dominant, that is, in the input image having a portion (edge) where the signal value changes rapidly.

  On the other hand, even if the first quantization matrix is transmitted in the case of using the TS mode in a block other than the 4 × 4 conversion block in the extension profile (profile other than the general-purpose profile) from such a problem, It is shown that quantization is performed using the second quantization matrix if the block is encoded by the TS mode. However, even in the extension profile, in the 4 × 4 transform block, for backward compatibility in implementation, whether the flag applying the first quantization matrix is valid regardless of whether it is in the TS mode or not. Even in the TS mode, there is a problem that the first quantization matrix is applied.

  The present invention has been made in view of the above points, and provides an image coding apparatus, an image decoding apparatus, an image coding program, and an image decoding program that can suppress deterioration in image quality due to quantization distortion. The purpose is to

  An image coding apparatus according to a first feature includes: a transform unit that performs orthogonal transform on a transform block that is a part of a residual image that is a residual between a predicted image and an input image; and 4 having a 4 × 4 size An enabling determination unit that determines whether or not to enable the conversion skip (TS mode) for skipping the orthogonal conversion for a × 4 conversion block in units of a sequence of one or a plurality of frames; For each 4 × 4 transform block corresponding to the sequence for which it is determined that the transform skip has been enabled by the second step, the second step comprises equal elements without using the first quantization matrix of different value elements And a quantization unit that performs quantization using a quantization matrix.

  An image decoding apparatus according to a second feature includes: an inverse quantization unit that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image; Inverse transform unit for performing inverse orthogonal transform on the transform block, and whether or not transform skip for skipping orthogonal transform is enabled for 4 × 4 transform block having a size of 4 × 4. And a validation determination unit that makes a determination in units of a sequence of frames. The inverse transform unit is a first quantization matrix including elements of different values with respect to each 4 × 4 transform block corresponding to the sequence determined to be enabled by the validation determination unit. Inverse quantization is performed using a second quantization matrix consisting of elements of equal values without using.

  An image coding apparatus according to a third feature performs a transform unit that performs orthogonal transform on a transform block that is a part of a residual image that is a residual between a predicted image and an input image, and performs quantization on the transform block. Quantizing one of a first quantization matrix consisting of elements of different values and a second quantization matrix consisting of elements of equal values, for a 4 × 4 transform block having a size of 4 × 4 And a quantization matrix determination unit that determines whether or not to use for each frame. The conversion unit is configured to perform a conversion skip for skipping the orthogonal conversion for each 4 × 4 conversion block corresponding to a frame determined to use the first quantization matrix for quantization by the quantization matrix determination unit. The orthogonal transform is performed on the 4 × 4 transform block without applying.

  An image decoding apparatus according to a fourth feature includes: an inverse quantization unit that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image; Performing inverse orthogonal transformation on the transformed block, and for a 4 × 4 transform block having a size of 4 × 4, a first quantization matrix composed of elements of different values and a second composed of elements of equal values And a quantization matrix determination unit that determines which one of the quantization matrices is to be used for inverse quantization in units of frames. The inverse transformation unit is a transformation that skips the inverse orthogonal transformation for each 4 × 4 transformation block corresponding to a frame determined to use the first quantization matrix for quantization by the quantization matrix determination unit. Inverse orthogonal transform is performed on the 4 × 4 transform block without applying skips.

  According to the present invention, an image coding apparatus, an image decoding apparatus, an image coding program, and an image decoding program capable of suppressing deterioration in image quality caused by quantization distortion are provided.

It is a block diagram which shows the structure of the image coding apparatus concerning 1st and 2nd embodiment. It is a figure which shows the example of a quantization matrix. It is a figure which shows an example of division of a conversion block. It is a flowchart which shows an example of the encoding process concerning 1st Embodiment. It is a block diagram which shows the structure of the image decoding apparatus which concerns on 1st and 2nd embodiment. It is a flowchart which shows an example of the decoding process concerning 1st Embodiment. It is a flowchart which shows an example of the encoding process which concerns on 2nd Embodiment. It is a flowchart which shows an example of the decoding process concerning 2nd Embodiment.

  Hereinafter, first and second embodiments of the present invention will be described with reference to the drawings. In the first and second embodiments, the image coding apparatus performs coding processing based on the HEVC scheme, and the image decoding apparatus performs decoding processing based on the HEVC scheme.

(1) First Embodiment (1.1) Image Coding Device An image coding device according to the first embodiment will be described.

(1.1.1) Configuration of Image Encoding Device FIG. 1 shows a configuration of an image encoding device 1 according to the first embodiment.

  As shown in FIG. 1, the image encoding device 1 includes a frame buffer 101, a subtraction unit 102, an encoded data generation unit 103, a control unit 109, an addition unit (synthesis unit) 110, a deblocking filter 111, a reference memory 112, The prediction unit 113, the lossless encoding unit 114, the accumulation buffer 115, and the motion vector calculation unit 116 are provided.

  The image coding apparatus 1 according to the first embodiment includes a transform unit 104 that performs orthogonal transform on a transform block that is a part of a residual image that is a residual between a predicted image and an input image, and quantization on the transform block. The quantization unit 105 for performing, and for a 4 × 4 transform block having a size of 4 × 4, either one of a first quantization matrix consisting of elements of different values and a second quantization matrix consisting of elements of equal values And a quantization matrix determination unit 109d that determines whether or not to use for quantization on a frame basis. The transformation unit 104 applies transformation skip for skipping orthogonal transformation to each 4 × 4 transformation block corresponding to a frame determined to be used for quantization by the quantization matrix determination unit 109 d. Without performing orthogonal transform on the 4 × 4 transform block.

  The frame buffer 101 temporarily stores, for each frame, an image signal input from the outside of the image encoding device 1.

  Note that, when storing the input image signal, the frame buffer 101 may rearrange the order of each frame according to a predetermined GOP (Group of Picture) structure. The GOP structure is an array of frame types (for example, intra frames, inter frames, etc.) consisting of a predetermined number of frames. An intra frame refers to a frame of an image signal to be encoded by performing intra-frame prediction (intra prediction). The inter frame refers to a frame of an image signal to be encoded by performing inter prediction.

  The frame buffer 101 identifies an image signal of a frame (also referred to as a current frame or current frame) to be currently processed from the stored input image signal, and performs a current process as a part of the image signal of the identified frame An input image block signal indicating an image of a current block or a current block) is read out.

  The order in which the frame buffer 101 selects the current block may be, for example, the same as the order of raster scan. The frame buffer 101 outputs the read input image block signal to the subtraction unit 102 and the motion vector calculation unit 116.

  The subtraction unit 102 subtracts the predicted image block signal input from the prediction unit 113 from the input image block signal input from the frame buffer 101 to generate a difference image block signal.

  Here, the subtraction unit 102 subtracts the signal value of the corresponding pixel among the signal values forming the reference image block signal from the signal value of each pixel forming the input image block signal, and generates the generated difference image block signal. It is output to the encoded data generation unit 103.

  The encoded data generation unit 103 generates encoded data in which the information amount is further reduced for the difference image block signal input from the subtraction unit 102, and controls the control unit 109 among the generated encoded data. The encoded data corresponding to the processing mode (mode) indicated by the signal is output to the lossless encoding unit 114. The processing mode indicated by the control signal includes, for example, determination as to whether or not to skip orthogonal transformation, orthogonal transformation, setting of a quantization matrix used when performing quantization, and quantization.

  Also, the encoded data generation unit 103 includes reverse processing, decodes the generated encoded data to generate a restored difference image block signal, and outputs the generated restored difference image block signal to the addition unit 110.

  The encoded data generation unit 103 includes a conversion unit 104, a quantization unit 105, an inverse quantization unit 106, an inverse conversion unit 107, and a quantization matrix setting unit 108.

  The transform unit 104 performs orthogonal transform on the differential image block signal input from the subtraction unit 102, generates orthogonal transform coefficient data, and outputs the generated transform coefficient data to the quantization unit 105.

  Orthogonal transformation means, for example, discrete cosine transform (DCT), discrete sine transform (DST) or the like. Note that, depending on the processing mode indicated by the control signal, the transform unit 104 may skip orthogonal transformation. In that case, the conversion unit 104 outputs the difference image block signal to the quantization unit 105 as conversion coefficient data without converting it. The operation mode that skips the orthogonal transformation may be referred to as a TS (Transform Skip) mode.

  The quantization unit 105 quantizes the conversion coefficient indicated by the conversion coefficient data input from the conversion unit 104 using the quantization parameter qP and the quantization matrix, and generates quantization conversion coefficient data indicating the quantized conversion coefficient. . The quantization parameter qP is commonly applied to the data in the block. The quantization matrix setting unit 108 sets a quantization matrix configured by quantization values specified for each transform coefficient in the block. The value (quantization value) of each element of the quantization matrix indicates the quantization width for quantizing the transform coefficient of the corresponding element of the transform coefficient data. The elements constituting the quantization matrix are sometimes called scaling factors, and examples of this quantization matrix will be described later. When orthogonal transformation is skipped, transform coefficient data input to the quantization unit 105 is a differential image block signal indicating the signal value of the differential image block. The quantization unit 105 outputs the generated quantization transformation coefficient data to the inverse quantization unit 106, and the quantization information (quantization parameter qP and quantization matrix) and the quantization transformation coefficient data are one of encoded data. Output to the lossless encoding unit 114 as a part.

  The inverse quantization unit 106 dequantizes the quantization conversion coefficient indicated by the quantization conversion coefficient data input from the quantization unit 105 using the quantization matrix set by the quantization matrix setting unit 108 and the quantization parameter qP. Quantize and calculate restoration transform coefficients. The quantization matrix used by the inverse quantization unit 106 is the same as the quantization matrix used when generating the quantization matrix. The inverse quantization unit 106 outputs restoration transform coefficient data indicating the calculated restoration transform coefficient to the inverse transform unit 107.

  The inverse transform unit 107 performs inverse orthogonal transform on the decompressed transform coefficient indicated by the decompressed transform coefficient data input from the inverse quantization unit 106, and generates a decompressed difference image block signal. The inverse transform unit 107 outputs the generated restored difference image block signal to the addition unit 110. The inverse orthogonal transform used by the inverse transform unit 107 is an inverse operation of the orthogonal transform performed by the transform unit 104. For example, when the transform unit 104 uses a DCT, the inverse transform unit 107 uses IDCT (Inverse DCT, inverse discrete cosine transform). When the transform unit 104 skips the orthogonal transform, the inverse transform unit 107 skips the inverse orthogonal transform. In that case, since the input restored transform coefficient data is a signal indicating the signal value of the quantized difference image block, the inverse transform unit 107 restores the input restored transform coefficient data. Output to the addition unit 110 as

  The quantization matrix setting unit 108 sets a quantization matrix to be used for quantization and dequantization in the quantization unit 105 and the dequantization unit 106 based on the processing mode indicated by the control signal input from the control unit 109. A plurality (for example, three) of candidate matrix data, which are candidates for the quantization matrix to be set, are stored in advance. One of the plurality of candidate matrix data is candidate matrix data of an initial value (default). The initial value candidate matrix data has, for example, a quantization value (e.g., 16) common to the elements. Hereinafter, candidate matrix data of quantization values common to the elements or the quantization matrix may be referred to simply as an initial value or a flat initial value. The quantization matrix consisting of elements of equal value corresponds to the second quantization matrix. The quantization matrices indicated by the remaining candidate matrix data may have different quantization values among the elements. The quantization matrix composed of elements of different values corresponds to the first quantization matrix. Here, the elements having different values are not limited to the case where the values of all the elements constituting the quantization matrix are different, and the values of some elements constituting the quantization matrix may be different. For example, in the quantization matrix, a plurality of groups each of which includes at least one element may be defined, the values of the elements may be equal in the groups, and the values of the elements may be different between the groups. The number of pieces of candidate matrix data indicating the second quantization matrix stored in the quantization matrix setting unit 108 is not limited to one, and may be two or more. In the following description, the first quantization matrix set by the quantization matrix setting unit 108 may be referred to as “set quantization matrix” or “non-flat quantization matrix”. The quantization matrix setting unit 108 is set in a predetermined manner in accordance with the type of the current frame (for example, intra frame, inter frame, etc.), the block size, and the type of signal value (for example, luminance signal, color difference signal, etc.). The quantization matrix may be selected. Further, the quantization matrix setting unit 108 may select the quantization matrix in accordance with the frequency characteristic of each differential image block.

  The control unit 109 controls the overall operation of the image encoding device 1. The control unit 109 determines, for each differential image block, as a processing mode, whether or not to skip orthogonal transformation, and which quantization matrix (for example, a flat quantization matrix or a non-flat quantization matrix) to select Do. The control unit 109 calculates, for example, a cost value indicating the degree of approximation of the restored difference image block signal with the difference image block signal for each processing mode, and the difference image block having the highest coding efficiency based on the calculated cost value. Select a processing mode that approximates the signal. The index is, for example, SAD (Sum of Absolute Differences, absolute difference sum), SSD (Sum of Squared Differences, square difference sum), RD (Rate-Distortion) cost or the like. The control unit 109 outputs a control signal indicating the determined processing mode to the encoded data generation unit 103, and outputs a control signal indicating the determination result to the lossless encoding unit 114 as a part of encoded data.

  The control unit 109 includes a division control unit 109a, a TS validation determination unit 109b, a TS application determination unit 109c, and a quantization matrix determination unit 109d.

  The division control unit 109a divides the differential image block (which is also a unit for the inverse transformation unit 107 to perform inverse transformation processing), which is a unit to be encoded, and controls to generate a further subdivided transformation block. Do. A process of dividing a difference image block into a conversion block as an original block is called block division. Details of the block division will be described later. The division control unit 109a configures a block size determination unit that determines, for each conversion block, whether it is a 4 × 4 conversion block having a 4 × 4 block size.

  The TS validation determination unit 109b determines whether to validate a transform skip (TS: Transform Skip) that skips orthogonal transformation on a transform block in a frame unit or a sequence unit including a plurality of frames. In the first embodiment, it is assumed that the conversion skip is enabled. Details of the TS validation determination unit 109b will be described in the second embodiment.

  The TS application determination unit 109c determines, for each difference image block (transformation block), whether or not to skip the orthogonal transformation. The TS application determination unit 109c may determine whether or not to skip the orthogonal transformation only when the TS validation determination unit 109b determines to validate the conversion skip on a frame basis or a sequence basis. The TS application determination unit 109 c generates a control signal (for example, transform_skip_flag = 1) indicating that orthogonal transformation is skipped for a transform block determined to skip orthogonal transformation. transform_skip_flag = 1 indicates that a transform skip is applied to the current transform block. On the other hand, the TS application determination unit 109 c generates a control signal (for example, transform_skip_flag = 0) indicating that orthogonal transformation is not skipped for a transform block determined to not skip orthogonal transformation.

  The quantization matrix determination unit 109d determines which quantization matrix (for example, a flat quantization matrix or a non-flat quantization matrix) to select for each difference image block (transform block) having a different size. For example, the quantization matrix determination unit 109d determines, for each 4 × 4 transform block, which of the flat quantization matrix and the non-flat quantization matrix is to be used for inverse quantization. As a flat quantization matrix, for example, there is an initial value designed such that all quantization widths defined in HEVC are 16 and so on. If it is determined that a non-flat quantization matrix is to be used, the quantization matrix determination unit 109d generates a control signal (for example, scaling_list_enable_flag = 1) indicating that the non-flat quantization matrix is to be used. scaling_list_enable_flag = 1 indicates that the scaling list is used for quantization of transform coefficients. The scaling list is a list showing each element (scaling factor) that constitutes a non-flat quantization matrix. Alternatively, the control signal indicating that a non-flat quantization matrix is to be used is a 4 × 4 block scaling_list_pred_mode_flag = 1 and indicates a non-flat quantization matrix or indicates that an encoder-defined quantization matrix is not flat. May be scaling_list_pred_mode_flag = 1 indicates that the value in the scaling list is explicitly indicated (notified). On the other hand, when it is determined that a flat quantization matrix is to be used, the quantization matrix determination unit 109d generates a control signal (for example, scaling_list_enable_flag = 0) indicating that a flat quantization matrix is to be used. scaling_list_enable_flag = 0 indicates that the scaling list is not used for quantization of transform coefficients. Alternatively, the control signal indicating that a flat quantization matrix is used is scaling_list_enable_flag = 1, 4 × 4 block scaling_list_pred_mode_flag = 1, and a flat quantization matrix is indicated or the encoder definition quantization matrix is flat. It may indicate that the

  In the first embodiment, the determination of the quantization matrix determination unit 109d is performed before the determination of the TS application determination unit 109c. The TS application determination unit 109c determines whether or not to apply conversion skip to a 4 × 4 transform block corresponding to a frame determined to use the first quantization matrix (non-flat quantization matrix) for quantization. I omit it. By omitting the determination of the TS application determination unit 109c, the amount of processing of the encoding process can be reduced and the encoding process can be speeded up. Then, the conversion unit 104 does not apply conversion skip to the 4 × 4 conversion block determined to use the first quantization matrix for quantization based on the control signal, and the 4 × 4 conversion block Perform orthogonal transformation and quantization processing for In other words, the transform unit 104 always performs orthogonal transform on a 4 × 4 transform block that uses the first quantization matrix for quantization. As a result, it is possible to prevent the conversion skip from being applied to the 4 × 4 conversion block to which the first quantization matrix is applied, which results in quantization distortion at the time of conversion skip of the 4 × 4 conversion block in the HEVC scheme. It is possible to suppress the deterioration of the image quality.

  The addition unit 110 adds (combines) the restored difference image block signal input from the inverse conversion unit 107 and the predicted image block signal input from the prediction unit 113 to generate a restored image block signal. Here, the adding unit 110 adds the signal value of each pixel forming the restored difference image block signal and the signal value of the corresponding pixel among the signal values forming the predicted image block signal. The addition unit 110 outputs the generated restored image block signal to the deblocking filter 111.

  The deblocking filter 111 performs filtering processing on the restored image block signal input from the adding unit 110 to remove the block distortion component. In addition, filter processing for improving sharpness (for example, sample adaptive filter: SAO) may be involved. The deblocking filter 111 stores the restored image block signal from which the components of the coding distortion have been removed in the reference memory 112 as a reference image block signal.

  The reference memory 112 stores the reference image block signal input from the deblocking filter 111 in a storage area corresponding to the display area corresponding to the reference image block for each frame, and the reference image signal indicating the reference image is a frame Remember each time.

  The prediction unit 113 performs known prediction processing using the reference image signal already stored in the reference memory 112 to generate a prediction image block signal according to the current block, and outputs the generated prediction image block signal to the subtraction unit 102. Do. Here, when the current frame is an intra frame, the prediction unit 113 performs intra prediction using a reference image signal related to the current frame to generate a predicted image block signal related to the current block. When the current frame is an inter frame, the prediction unit 113 performs inter prediction using a reference image signal related to a frame different from the current frame to obtain a predicted image block signal related to the current block. Here, the prediction unit 113 extracts a block at a position designated by the motion vector input from the motion vector calculation unit 116 from the reference image signal with reference to the current block as a predicted image block signal. The frame of the reference image signal used for inter prediction is determined by which frame the current frame is in the frame order according to the predetermined GOP structure. The prediction unit 113 outputs the mode data indicating the selected prediction mode to the lossless encoding unit 114 as a part of the encoded data.

  The motion vector calculation unit 116 compares reference image data indicating a frame of a reference image signal to be referred to when the prediction unit 113 generates a predicted image block signal with the input image block signal input from the frame buffer 101, Calculate motion vector. The motion vector calculation unit 116 determines a vector from the representative point (for example, the pixel at the upper left end) of the current block to the representative point of the selected image block as a motion vector. The motion vector calculation unit 116 includes a storage unit (not shown) that stores a predetermined motion vector for a predetermined period, and is a candidate derived from the motion vectors of blocks near the current block having a small distance to the predetermined motion vector. A motion vector is selected as a prediction vector (vector prediction). The motion vector calculation unit 116 calculates a difference vector between the determined motion vector and the selected prediction vector, and the lossless encoding unit uses the calculated difference vector and a vector index indicating the determined prediction vector as part of the encoded data. Output to 114. The motion vector calculation unit 116 outputs the determined motion vector to the prediction unit 113.

  The lossless encoding unit 114 receives, as part of the encoded data, the quantized transform coefficient data, the candidate matrix data, or the code (if applicable) relating to the data from the encoded data generation unit 103, The control signal is input as a part of the encoded data. The lossless encoding unit 114 receives mode data from the prediction unit 113 as a part of encoded data, and receives a difference vector and a vector index (if applicable) from the motion vector calculation unit 116. The lossless encoding unit 114 integrates the input encoded data for each frame, and performs lossless encoding, eg, entropy encoding, on the integrated data (integrated data) to generate a bit stream. The lossless encoding unit 114 stores the generated bit stream in the accumulation buffer 115.

  The accumulation buffer 115 stores the bit stream input from the lossless encoding unit 114. The accumulation buffer 115 sequentially outputs the stored bit stream to the outside of the image coding device 1, for example, to the image decoding device 2.

(1.1.2) Example of quantization matrix Next, an example of the quantization matrix will be described. FIG. 2 is a diagram illustrating an example of the quantization matrix.

  FIGS. 2A and 2B show quantization matrices tb11 and tb21 each having 16 quantization values in total (four columns in the horizontal direction and four columns in the vertical direction). The squares included in the quantization matrices tb11 and tb21 indicate elements. The numbers described in each of the squares indicate the quantization value, which is also represented by the shading of the square area. In the example shown in FIG. 2, the quantization value is larger as the area is filled in darker, and the quantization value is smaller as the area is thinner.

  As shown in FIG. 2A, the quantization matrix tb11 is an example of a flat quantization matrix (second quantization matrix), and the quantization value of each element is 16. By using such a quantization matrix, it is possible to quantize the transform coefficient to be quantized and the signal value with the same precision regardless of the elements. Therefore, when the orthogonal transformation is not performed, it is possible to avoid the deterioration of the image quality by using the quantization matrix having a bias in the quantization value such as the quantization matrix tb21 and the like.

  As shown in FIG. 2B, the quantization matrix tb21 is an example of a non-flat quantization matrix (first quantization matrix). The quantization matrix tb 21 has a larger quantization value as the order in the horizontal direction and the vertical direction increases. The quantization values of the upper left end (the first row and the first column), the middle (the third row and the second column), and the lower right end (the fourth row and the fourth column) of the quantization matrix tb21 are 16, 22, and 42, respectively. . In such a quantization matrix, orthogonal transform coefficients associated with elements arranged at the upper left, that is, lower-range transform coefficients, are quantized with high accuracy, and conversion coefficients associated with the elements arranged at the lower right, ie, higher The domain transform coefficients are quantized with lower precision. Therefore, when orthogonal transformation is performed, the amount of information in the high region can be obtained by quantization without degrading the image quality by utilizing the human visual characteristic that the lower region is more sensitive to the spatial change in density and hue. It can be reduced. On the other hand, in the case where orthogonal transformation is not involved, different distortions occur in each coordinate, and the deterioration at the lower right of the block becomes remarkable.

  The quantization matrix setting unit 108 selects a quantization matrix having the same number of elements as the block size when the number of elements of the difference image block, that is, the block size is variable. In that case, the quantization matrix setting unit 108 acquires a quantization matrix corresponding to each of the candidate block sizes. The candidate block sizes are, for example, 4 in the horizontal direction, 4 in the vertical direction (16 in total), 8 in the horizontal direction, 8 in the vertical direction (64 in total), 16 in the horizontal direction, 16 in the vertical direction (total (256 pieces), 32 pieces in the horizontal direction, 32 pieces in the vertical direction (1024 pieces in total), etc.

(1.1.3) Block Division Next, block division will be described.

  The division control unit 109a of the control unit 109 divides the difference image block (which is also a unit for the inverse conversion unit 107 to perform inverse conversion processing), which is a unit to be encoded, and generates a further divided conversion block. Control the The process of dividing this difference image block into the conversion block as the original block is called block division. A differential image block which is a unit to be coded may be called a coding unit (CU), and a transform block may be called a transform unit (TU). The division control unit 109a performs, for example, block division described in Non-Patent Document 1. Thus, the block size of the transform block is variable. In that case, the size (number of elements) of the quantization matrix used in quantization and dequantization is equal to the block size of the transform block.

  Here, when the block size of the original block is a predetermined block size (for example, 64 in the horizontal direction and 64 in the vertical direction), the division control unit 109a horizontally or vertically divides the original block (layer 0). Attempt to divide into four blocks (first hierarchy) each having half the number of elements in the direction. The division control unit 109a determines whether or not division is to be performed on the basis of whether or not an index (for example, RD cost) indicating the size of the residual decreases due to division. Specifically, when the index for a certain block is larger than the total value of the indices for each block obtained by temporarily dividing the block, the division control unit 109a determines that the block is to be divided.

  Therefore, the division control unit 109a temporarily divides each of the blocks in a certain hierarchy into four blocks in the next hierarchy, and determines whether or not to divide the number into a predetermined number of hierarchies (for example, four hierarchies). Repeat until it reaches The division control unit 109a outputs division information indicating the division of the divided transform block as a control signal to the lossless encoding unit 114 and the encoded data generation unit 103. The division information indicating the division of the conversion block can be regarded as information indicating the block size of the conversion block.

  FIG. 3 is a diagram showing an example of division of transform blocks. In FIG. 3, an original block tb31 is divided into three layers. Blocks tb317, tb318, and tb319 on the upper right, lower left, and lower right of the original block tb31 are conversion blocks of the first hierarchy. The upper left region of the original block tb31 is further divided. The upper right, lower left and lower right blocks tb 314, tb 315 and tb 316 in the area are conversion blocks of the second hierarchy. The upper left area of the area is further divided. The divided regions are four transform blocks tb310, tb311, tb312, and tb313, which are the transform blocks of the third layer.

(1.1.4) Operation of Encoding Process Next, an example of the encoding process according to the first embodiment will be described. FIG. 4 is a flowchart showing an example of the encoding process according to the first embodiment. The flow shown in FIG. 4 shows the encoding process for each transform block constituting the differential image block signal, and is applied to the transform block determined to be a 4 × 4 transform block by the division control unit 109a.

  As shown in FIG. 4, in step S101, whether the quantization matrix determination unit 109d uses a flat quantization matrix (second quantization matrix) or a non-flat quantization matrix (first quantization matrix)? Determine If it is determined that a flat quantization matrix is to be used (step S101: NO), the quantization matrix determination unit 109d generates a control signal indicating that a flat quantization matrix is to be used. Output to the part 114. Thereafter, the process proceeds to step S102.

  In step S102, the quantization matrix setting unit 108 quantizes the candidate matrix data representing the predetermined flat quantization matrix based on the control signal indicating that the flat quantization matrix is used, and dequantizes the candidate matrix data. It is set in the part 106. The quantization value m [x] [y] indicated by the candidate matrix data is, for example, indicated by equation (1).

m [x] [y] = 16 (1)
Here, x and y are integers indicating elements in the horizontal direction and the vertical direction, respectively, and take values from 0 to xTbS-1 and from 0 to yTbS-1. xTbS and yTbS are integers (for example, any of 4, 8, 16, and 32) indicating the numbers of elements in the horizontal direction and the vertical direction of the conversion block, respectively. In the first embodiment, xTbS and yTbS are both 4 because 4 × 4 conversion blocks are targeted. Thereafter, the process proceeds to step S103.

  In step S103, the TS application determination unit 109c determines whether to skip the conversion process (orthogonal transformation). Specifically, the TS application determination unit 109 c determines whether to apply the TS mode to the 4 × 4 conversion block to be processed. If it is determined that the conversion process is to be skipped (step S103: YES), the TS application determination unit 109c transmits a control signal (transform_skip_flag = 1) indicating that the conversion process is skipped to the encoded data generation unit 103 and the lossless encoding unit Output to 114. Thereafter, the process proceeds to step S107. On the other hand, when it is determined that the conversion process is not skipped (step S103: NO), the TS application determination unit 109c converts the control signal (transform_skip_flag = 0) indicating that the conversion process is not skipped to the encoded data generation unit 103 and lossless It is output to the encoding unit 114.

  Thereafter, in step S104, the transform unit 104 performs orthogonal transform on the 4 × 4 transform block to be processed based on the control signal indicating that the transform process is not skipped. Then, the process proceeds to step S107.

  On the other hand, when it is determined to use a non-flat quantization matrix (step S101: YES), the quantization matrix determination unit 109d does not skip the conversion processing and a control signal indicating that the non-flat quantization matrix is used. The control signal (transform_skip_flag = 0) shown is output to the encoded data generation unit 103 and the lossless encoding unit 114. The quantization matrix determination unit 109d may instruct the TS application determination unit 109c to output a control signal indicating not to skip the conversion process. Here, if it is determined that a non-flat quantization matrix is to be used, the TS application determination unit 109c does not determine whether to skip the conversion process.

  In step S105, the quantization matrix setting unit 108 sets the non-flat quantization matrix in the quantization unit 105 and the inverse quantization unit 106 based on the control signal indicating that the non-flat quantization matrix is used.

  Then, in step S106, the transform unit 104 performs orthogonal transform on the 4 × 4 transform block to be processed based on the control signal indicating that the transform process is not skipped. Then, it progresses to S107.

  In step S107, the quantization unit 105 quantizes the 4 × 4 transform block to be processed using the quantization matrix set by the quantization matrix setting unit 108 and the quantization parameter qP to generate quantization coefficient data. . The quantization unit 105 outputs the generated quantization coefficient data to the inverse quantization unit 106 and the lossless encoding unit 114. When the quantization unit 105 performs quantization, for example, a transform coefficient level value TransCoeffLevel [xTbY] [yTbY] [cIdx] at which the value on the right side of Equation (2) most approximates to the transform coefficient d [x] [y]. Select [x] [y].

d [x] [y] = Clip3 (coeffMin, coeffMax, ((TransCoeffLevel [xTbY] [yTbY] [cIdx] [x] [m] * m [x] [y] * levelScale [qP% 6] << ( qP / 6)) + (1 << (bdShift-1))) >> bdShift) (2) (H. 265 8-311)
Here, Clip 3 (a, b, xx) is defined as a when the real number xx is smaller than the real number a, and as b when the real number xx is larger than the real number b, and the real number xx is a or When it is larger than a and smaller than b or b, it is a function that determines as it is as it is. xTbY and yTbY indicate horizontal and vertical coordinate values of the upper left end of the conversion block (target block) to be processed. coeffMin and coeffMax respectively indicate the minimum value and the maximum value of the signal value of each element indicated by 16 bits. cIdx is an index indicating the type of signal value. cIdx = 0, 1, 2 respectively indicate a luminance signal, a color difference signal Cb, and a color difference signal Cr. levelScale [0] -levelScale [5] (following H.265 8.4.4.2.7 Decoding process for palette mode (8-76)) are 40, 45, 57, 64 and 72, respectively. . qP is a quantization parameter, that is, an integer indicating a quantization accuracy, and a parameter that halves a quantization value every six increments. qP% 6 shows the remainder obtained by dividing qP by six. a << b is a bit shift operator indicating that the value of a is shifted to the left by b digits in binary notation, that is, multiplied by 2 to the power of b. a >> b is a bit shift operator indicating that the value of a is shifted to the right by b digits in binary notation, that is, division of 2 to the power of b is performed. bdShift is a bit shift value predetermined according to the type of signal value. For example, when the signal value is a luminance signal Y, BdShift _is BitDepth Y + Log2 (nTbS) -5 . Bit Depth _Y indicates the bit depth of the luminance signal Y, that is, the number of quantization bits (for example, 10 bits). nTbS indicates the block size of the target block. If the signal value is a color difference signal Cb, Cr, bdShift _is BitDepth C + Log2 (nTbS) -5 . Bit Depth _C indicates the bit depth of the luminance signals Cb and Cr, that is, the number of quantization bits (for example, 10 bits). Here, qP may have different values depending on the type of frame. Thereafter, the process proceeds to step S108.

  In step S108, the quantization unit 105 outputs quantized transform coefficient data indicating the selected coefficient level value to the lossless encoding unit 114 as a part of the encoded data.

  In addition, the quantization unit 105 outputs the quantization transformation coefficient data to the inverse quantization unit 106. The output quantization transformation coefficient data is inverse-quantized in the inverse quantization unit 106, and the quantized transformation coefficient d [x] [y] is calculated. Thereafter, the process shown in FIG. 4 is ended.

  As described above, when using a non-flat quantization matrix in the 4 × 4 transform block, it is possible to avoid the determination of application / non-application of the TS mode, and high speed and low power consumption can be realized. Moreover, when using a non-flat quantization matrix, generation | occurrence | production of the spatial quantization distortion by selecting TS mode can be suppressed, and the subjective quality of a decoded image can be improved.

(1.2) Image Decoding Device An image coding device according to the first embodiment will be described.

(1.2.1) Configuration of Image Decoding Device FIG. 5 is a block diagram showing a configuration of the image decoding device 2 according to the first embodiment.

  As shown in FIG. 5, the image decoding apparatus 2 includes a control unit 200, an accumulation buffer 201, a lossless decoding unit 202, an encoded data decoding unit 203, an addition unit (composition unit) 207, a deblocking filter 208, a reference memory 209, A prediction unit 210, a frame buffer 211, and a motion vector calculation unit 212 are provided.

  The image decoding apparatus 2 according to the first embodiment includes an inverse quantization unit 204 that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image; An inverse transform unit 205 for performing inverse orthogonal transform on the transform block applied, and a first quantization matrix composed of elements of different values and a first transform matrix composed of elements of equal values for a 4 × 4 transform block having a size of 4 × 4. And a quantization matrix determination unit 200d that determines which one of the two quantization matrices is to be used for inverse quantization in units of frames. The inverse transformation unit 205 performs transformation skip for skipping inverse orthogonal transformation with respect to each 4 × 4 transformation block corresponding to a frame determined to be used for quantization by the quantization matrix determination unit 200 d. Inverse orthogonal transform is performed on the 4 × 4 transform block without application.

  The bit stream is input to the accumulation buffer 201 from the outside of the image decoding device 2, for example, the image coding device 1, and the input bit stream is temporarily stored. In the accumulation buffer 201, the input bit stream is integrated for each frame.

  The lossless code decoding unit 202 reads a bit stream for each frame from the accumulation buffer 201, performs lossless code decoding on the read bit stream, and generates coded data. The lossless code decoding method used by the lossless code decoding unit 202 is a decoding method corresponding to the lossless coding method used by the lossless coding unit 114.

  The lossless code decoding unit 202 outputs the quantization conversion coefficient data, the candidate matrix data, or the code (if applicable) related to the data to the coded data decoding unit 203 among the decoded coded data. The lossless code decoding unit 202 outputs a control signal to the control unit 200 among the decoded encoded data. The lossless code decoding unit 202 outputs mode data of the decoded encoded data to the prediction unit 210, and outputs a difference vector and a vector index (if applicable) to the motion vector calculation unit 212.

  The encoded data decoding unit 203 includes an inverse quantization unit 204, an inverse transform unit 205, and a quantization matrix setting unit 206.

  The inverse quantization unit 204 inversely quantizes the quantization conversion coefficient indicated by the quantization conversion coefficient data input from the lossless code decoding unit 202 using the quantization matrix set by the quantization matrix setting unit 206 and converts the conversion coefficient Calculate The inverse quantization unit 204 outputs restoration transform coefficient data indicating the calculated transform coefficient to the inverse transform unit 205.

  The inverse transform unit 205 performs inverse orthogonal transform on the restored transform coefficients input from the inverse quantization unit 204 to generate a restored difference block image signal. The inverse transform unit 205 outputs the generated restored difference block image signal to the addition unit 207. The inverse orthogonal transform used by the inverse transform unit 205 is an inverse operation of the orthogonal transform performed by the transform unit 104 (see FIG. 1). In addition, when the control signal input from the control unit 200 indicates to skip the inverse orthogonal transformation, the inverse transformation unit 205 skips the inverse orthogonal transformation. In that case, the input quantized transform coefficient data is a signal indicating the signal value of the quantized difference block image. Therefore, the inverse conversion unit 205 outputs the input restoration conversion coefficient data as a restoration difference block image signal to the addition unit 207 without performing inverse conversion.

  The quantization matrix setting unit 206 dequantizes the candidate matrix data corresponding to the processing mode indicated by the control signal input from the control unit 200 and the lossless code decoding unit 202 or the quantization matrix indicated by the code relating to the data. Set to The quantization matrix setting unit 206 has a configuration for acquiring the same candidate matrix data as the quantization matrix setting unit 108 (FIG. 1), for example, a storage unit.

  The control unit 200 controls the overall operation of the image decoding apparatus 2 based on the control signal output from the lossless code decoding unit 202. The control unit 200 includes a block size determination unit 200a, a TS validation determination unit 200b, a TS application determination unit 200c, and a quantization matrix determination unit 200d.

  The block size determination unit 200a determines, for each conversion block, whether it is a 4 × 4 conversion block having a 4 × 4 block size, and outputs a control signal indicating the determination result to the encoded data decoding unit 203. . For example, the block size determination unit 200a determines whether or not it is a 4 × 4 conversion block based on the division information indicating the division of the divided conversion block in the control signal.

  Whether or not the conversion skip is enabled in frame units or sequence units based on a control signal (for example, transform_skip_enabled_flag) indicating whether or not the conversion skip is enabled in frame units or sequence units. It is determined whether or not it is not, and a control signal indicating the determination result is output to the encoded data decoding unit 203. In the first embodiment, it is assumed that the conversion skip is enabled. Details of the TS validation determination unit 200b will be described in the second embodiment.

  The TS application determination unit 200 c determines, for each transform block, whether to skip the inverse orthogonal transformation based on the control signal (transform_skip_flag) indicating whether to skip the orthogonal transformation, and indicates a control signal indicating the determination result. Are output to the encoded data decoding unit 203. For example, when transform_skip_flag = 1 for the transform block to be processed, the TS validation determination unit 200b determines to skip the inverse orthogonal transform. When transform_skip_flag = 0 for the transform block to be processed, the TS validation determination unit 200b may determine not to skip the inverse orthogonal transform.

  The quantization matrix determination unit 200d uses a flat quantization matrix or does not use a flat quantization matrix for each transform block having different sizes based on a control signal indicating whether to use a flat quantization matrix or a non-flat quantization matrix. It is determined whether a non-flat quantization matrix is to be used for inverse quantization, and a control signal indicating the determination result is output to the encoded data decoding unit 203. For example, when scaling_list_enable_flag = 1 for the transform block to be processed, the quantization matrix determination unit 200 d determines that a non-flat quantization matrix is to be used for inverse quantization. Alternatively, if the quantization matrix determination unit 200 d indicates that the control signal indicates 4 × 4 block scaling_list_pred_mode_flag = 1 and the non-flat quantization matrix or indicates that the quantization matrix in the encoder definition is not flat, the non-flat quantum is determined. It may be determined that the quantization matrix is used for inverse quantization. On the other hand, when scaling_list_enable_flag = 0, the quantization matrix determination unit 200d determines that a flat quantization matrix is to be used for inverse quantization. Alternatively, the quantization matrix determination unit 200d indicates that the control signal is scaling_list_enable_flag = 1, indicates 4 × 4 block scaling_list_pred_mode_flag = 1, and indicates a flat quantization matrix or the quantization matrix in the encoder definition is flat. If it indicates, it may be determined that a flat quantization matrix is to be used for inverse quantization.

  In the first embodiment, the quantization matrix determination unit 200d performs the above-described determination before the determination of the TS application determination unit 200c. The TS application determination unit 200c omits the determination as to whether or not to apply the transform skip to the 4 × 4 transform block determined to use the non-flat quantization matrix (first quantization matrix) for the inverse quantization. As a result, the determination of the TS application determination unit 200c can be omitted, the processing amount of the decoding process can be reduced, and the decoding process can be speeded up.

  Then, the inverse transform unit 205 performs inverse orthogonal transform on the 4 × 4 transform block without applying a transform skip to the 4 × 4 transform block determined to use the non-flat quantization matrix for the inverse quantization. I do. In other words, the inverse transform unit 205 always performs inverse orthogonal transform on a 4 × 4 transform block that uses a non-flat quantization matrix for inverse quantization. As a result, it is possible to prevent the conversion skip from being applied to the 4 × 4 transform block to which the non-flat quantization matrix is applied, so that the image quality degradation due to the quantization distortion of the 4 × 4 transform block in the HEVC scheme It can be suppressed.

  The addition unit 207 adds (combines) the restored difference image block signal input from the inverse conversion unit 205 and the predicted image block signal input from the prediction unit 210 to generate a restored image block signal. The addition unit 207 outputs the generated restored image block signal to the deblocking filter 208.

  The deblocking filter 208 performs filtering processing on the restored image block signal input from the adding unit 207 to remove the block distortion component. In addition, filter processing (e.g., sample adaptive filter: SAO) for improving sharpness may be included, which represents the clarity of a minute portion of an image. The deblocking filter 208 stores the restored image block signal from which the component of block distortion has been removed as a reference image block signal in the reference memory 209 and the frame buffer 211.

  The reference memory 209 stores, for each frame, the reference image block signal input from the deblocking filter 208 in a storage area corresponding to the display area corresponding to the reference image block. Thereby, a reference image signal indicating a reference image can be stored for each frame.

  The prediction unit 210 generates a predicted image block signal according to the current block using the reference image signal stored in the reference memory 209, and outputs the generated predicted image block signal to the addition unit 207. If the current frame is an intra frame, the prediction unit 210 performs intra prediction. When the current frame is an inter frame, the prediction unit 210 performs inter prediction using a reference image signal related to a frame different from the current frame to generate a predicted image block signal related to the current block. The frame of the reference image signal used for inter prediction is determined by which frame the current frame is in the frame order according to the predetermined GOP structure. When performing inter prediction, the prediction unit 210 performs prediction processing in a prediction mode (for example, a motion compensation prediction mode, a merge mode, and the like) indicated by mode data input from the lossless code decoding unit 202.

  When performing inter prediction, the prediction unit 210 extracts a reference image block signal of a region designated by the motion vector input from the motion vector calculation unit 212 from the reference image signal of the determined frame, and extracts the extracted reference image block The signal is used to generate a predicted image block signal.

  The frame buffer 211 stores, for each frame, the restored image block signal input from the deblocking filter unit 208 in a storage area corresponding to a display area corresponding to the restored image block. Thereby, the frame buffer 211 stores the restored image signal for each frame in the storage area. The frame buffer 211 rearranges the restored image signals in the order of frames according to a predetermined GOP structure. The frame buffer 211 outputs the restored image signal to the outside of the image decoding device 2 for each frame, for example, to an image display device (display).

(1.2.2) Example of Decoding Process Next, an example of the decoding process according to the first embodiment will be described. FIG. 6 is a flowchart showing an example of the decoding process according to the first embodiment. In the flow shown in FIG. 6, the decoding process is performed for each conversion block. In addition, the flow illustrated in FIG. 6 is applied to a conversion block determined to be a 4 × 4 conversion block by the block size determination unit 200 a.

  As shown in FIG. 6, in step S201, the encoded data decoding unit 203 outputs a control signal, quantization conversion coefficient data, candidate matrix data or a code relating to that data from the lossless code decoding unit 202 as a part of the encoded data. Enter Thereafter, the process proceeds to step S202.

  In step S202, the quantization matrix determination unit 200d indicates that the input control signal indicates that the quantization matrix used for inverse quantization of the quantization conversion coefficient uses a flat quantization value or is not flat. This shows that the quantization matrix is used, and it is determined whether it is a matrix of quantized values flat with respect to 4 × 4. When it is determined that the flat quantization matrix is used (step S202: NO), the quantization matrix determination unit 200d outputs a control signal indicating that the flat quantization matrix is used to the encoded data decoding unit 203. Do. Thereafter, the process proceeds to step S203.

  In step S203, the quantization matrix setting unit 206 sets candidate matrix data indicating a predetermined flat quantization matrix in the inverse quantization unit 204 based on a control signal indicating that a flat quantization matrix is to be used. . The quantization value indicated by this candidate matrix data may be, for example, m [x] [y] indicated by equation (1). Next, in step S204, the inverse quantization unit 204 performs inverse quantization on the 4 × 4 transform block to be processed using a quantization matrix formed of flat quantization values. The inverse quantization unit 204 performs quantization on the candidate matrix data set by the quantization matrix setting unit 206 and the quantization parameter Qp, with the quantization conversion coefficient indicated by the quantization conversion coefficient data input from the lossless code decoding unit 202 Inverse quantize using values. Here, the inverse quantization unit 204 calculates, for example, the restoration transform coefficient d [x] [y] using Expression (2). In the process of calculating, the transform coefficient level value indicated by the quantized transform coefficient data TransCoeffLevel [xTbY] [yTbY] [cIdx] [x] [y] and the quantization value m [x] [y] indicated by the candidate matrix data The restoration conversion coefficient d [x] [y] may be calculated using this. Thereafter, the process proceeds to step S205.

  In step S205, the TS application determination unit 200c determines whether or not the input control signal indicates that the conversion process is to be skipped. If it is determined that the transformation process is to be skipped (transform_skip_flag = 1) (step S205: YES), the TS application determination unit 200c outputs a control signal indicating to skip the inverse orthogonal transformation. In this case, the inverse transform unit 205 skips inverse orthogonal transform on the 4 × 4 transform block to be processed, and outputs the restored transform coefficient data input from the inverse quantization unit 204 to the addition unit 207 as a restored difference image block. . On the other hand, when it is determined that the transformation processing is not skipped (transform_skip_flag = 0) (step S205: NO), the TS application determination unit 200c outputs a control signal indicating that the inverse orthogonal transformation is not skipped. And proceed to step S206.

  In step S206, the inverse transform unit 205 performs inverse orthogonal transform processing on the decompressed transform coefficient indicated by the decompressed transform coefficient data input from the dequantization unit 204 to generate a decompressed difference image block. The inverse transform unit 205 outputs the generated restored difference image block to the addition unit 207.

  On the other hand, when it is determined that the non-flat quantization matrix is to be used (step S202: YES), the quantization matrix determination unit 200d outputs a control signal indicating that the non-flat quantization matrix is to be used to the encoded data decoding unit 203. , And proceeds to step S207.

  In step S207, the quantization matrix setting unit 206 sets a quantization matrix indicated by the control signal input from the lossless code decoding unit 202 in the dequantization unit 204. Thereafter, the process proceeds to step S208.

  In step S208, the inverse quantization unit 204 performs inverse quantization on the 4 × 4 transform block to be processed using a non-flat quantization matrix.

  Next, in step S 209, the inverse transformation unit 205 performs inverse orthogonal transformation processing on the reconstruction transform coefficient indicated by the reconstruction transform coefficient data input from the inverse quantization unit 204 to generate a reconstruction difference image block, and the generated reconstruction The difference image block is output to the addition unit 207. When it is determined that a non-flat quantization matrix is to be used, the TS application determination unit 200c does not determine whether to skip the conversion process.

  As described above, since the TS mode is not used when a non-flat quantization matrix (first quantization matrix) is used, the TS application determination unit 200c can omit the determination as to whether or not orthogonal transform is applied. In addition, since the signal values of the respective elements of the transform block in which the orthogonal transform has not been performed are dequantized with equal precision, it is possible to avoid the deterioration of the image quality due to unequal precision in the block. For example, for an image in which a region including many high frequency components having high frequency components is distributed, the subjective quality of the decoded image can be improved.

(1.3) Summary of the First Embodiment In the image coding apparatus 1 according to the first embodiment, the conversion unit 104 is determined to use the first quantization matrix (non-flat quantization matrix) for quantization. The 4 × 4 transform block corresponding to the frame is subjected to orthogonal transform on the 4 × 4 transform block without applying a transform skip. In other words, the transform unit 104 always performs orthogonal transform on a 4 × 4 transform block that uses the first quantization matrix for quantization. In the image decoding apparatus 2 according to the first embodiment, the inverse transform unit 205 applies a transform skip to the 4 × 4 transform block determined to use the first quantization matrix for inverse quantization. Instead, inverse orthogonal transform is performed on the 4 × 4 transform block. In other words, the inverse transform unit 205 always performs inverse orthogonal transform on a 4 × 4 transform block that uses the first quantization matrix for inverse quantization. Therefore, the conversion skip can not be applied to the 4 × 4 transform block to which the first quantization matrix is applied, so that the image quality of a general image such as a natural image can be improved in the HEVC method.

  In the image encoding device 1 according to the first embodiment, the TS application determination unit 109b is a 4 × 4 transform block corresponding to a frame determined to use the second quantization matrix (flat quantization matrix) for quantization. Enable conversion skipping for. In other words, the TS application determination unit 109 b does not apply the conversion skip to the 4 × 4 transform block that uses the first quantization matrix for quantization, and uses the second quantization matrix for the 4 × 4 transform block that performs quantization. Make conversion skip applicable. Further, in the image decoding apparatus 2 according to the first embodiment, the TS application determination unit 200c performs 4 × 4 conversion corresponding to a frame determined to use the second quantization matrix (flat quantization matrix) for quantization. Enable conversion skipping for the block. In other words, the TS application determination unit 200c does not apply the conversion skip to the 4 × 4 transform block that uses the first quantization matrix for quantization, and uses the second quantization matrix for the 4 × 4 transform block that performs quantization. Make conversion skip applicable. In this way, by performing quantization and inverse quantization on the 4 × 4 transform block to which the transform skip is applied using the second quantization matrix, the image quality of CG and an image in which CG and a natural image are mixed are obtained. It can improve.

  Furthermore, in the image coding device 1 according to the first embodiment, the TS application determination unit 109 c is 4 × 4 corresponding to a frame determined to use the first quantization matrix (non-flat quantization matrix) for quantization. The determination as to whether or not to apply the conversion skip is omitted for the conversion block. In addition, in the image decoding device 2 according to the first embodiment, the TS application determination unit 200c performs the conversion skip on the 4 × 4 conversion block corresponding to the frame determined to use the first quantization matrix for inverse quantization. The determination of whether or not to apply is omitted. As a result, the processing amount of the encoding process and the decoding process can be reduced, and the encoding process and the decoding process can be speeded up.

(2) Second Embodiment Next, the difference between the second embodiment and the first embodiment will be mainly described.

(2.1) Image Encoding Device An image encoding device 1 according to the second embodiment will be described.

(2.1.1) Configuration of Image Encoding Device The functional block configuration of the image encoding device 1 according to the second embodiment is the functional block configuration of the image encoding device 1 according to the first embodiment (see FIG. 1). Is the same as However, the operation of some of the blocks is different from that of the first embodiment.

  The image coding apparatus 1 according to the second embodiment includes: a transform unit 104 which performs orthogonal transform on a transform block which is a part of a residual image which is a residual between a predicted image and an input image; TS validation determination unit 109d that determines whether or not to perform transform skip (TS mode) for skipping orthogonal transform for a 4 × 4 transform block that it has in units of a sequence of one or more frames, and TS valid For each 4 × 4 transform block corresponding to the sequence determined to be transform-skip enabled by the quantization determination unit 109d, without using a first quantization matrix consisting of elements of different values, from elements of equal values And a quantization unit 105 that performs quantization using the second quantization matrix. As described above, the image encoding device 1 according to the second embodiment performs 4 × 4 conversion as a quantization matrix common to the sequence level (SPS: Sequence parameter set) when enabling the TS mode in units of sequences. Set a flat quantization matrix (second quantization matrix) for the block.

  In the second embodiment, the TS validation determination unit 109b determines whether to enable conversion skipping (that is, whether or not the TS mode can be selected) in units of a sequence of one or a plurality of frames. Do. If it is determined that the conversion skip is enabled in units of sequences, the TS enable determination unit 109b outputs a control signal (for example, transform_skip_enabled_flag = 1) to that effect. transform_skip_enabled_flag = 1 indicates that transform_skip_flag may be present for the corresponding sequence. On the other hand, when it is determined that the conversion skip is not enabled in sequence units, the TS validation determination unit 109b outputs a control signal (for example, transform_skip_enabled_flag = 0) to that effect. transform_skip_enabled_flag = 0 indicates that transform_skip_flag does not exist for the corresponding sequence. The TS validation determination unit 109 b detects, for example, whether the input signal is CG (or a property similar to CG) in the pre-processing of encoding, and validates the TS mode. The TS validation determination unit 109b may analyze the already encoded video to validate the TS mode.

  In the second embodiment, the quantizing unit 105 generates a first quantum composed of elements of different values with respect to a 4 × 4 transform block in a frame making up the sequence for which the transform skip is enabled by the TS validation determination unit 109 b. It is set that quantization is performed using a second quantization matrix (flat quantization matrix) consisting of elements of equal values without using a quantization matrix (non-flat quantization matrix). In other words, the quantization unit 105 always performs the second quantization matrix when performing quantization on the 4 × 4 transform block for the frames that constitute the sequence for which the conversion skip is enabled by the TS validation determination unit 109 b. Use This makes it possible to prevent the first quantization matrix from being applied to the 4 × 4 transform block in which the orthogonal transform is skipped without setting in units of pictures.

  The quantization unit 105 uses the quantization information (that is, the encoded signal related to the quantization matrix) to be output to the lossless encoding unit 114 for the frame that constitutes the sequence for which the conversion skip is activated by the TS activation determination unit 109b. , And may be replaced by virtual data with a small amount of code to be replaced. As described later, the image decoding device 2 implicitly performs inverse quantization using a second quantization matrix (flat quantization matrix) consisting of elements of equal values for a sequence for which conversion skip is enabled. Do. In this case, the image decoding device 2 can ignore the coded signal related to the 4 × 4 block quantization matrix. By replacing the encoded signal relating to the quantization matrix with virtual data having a small code amount, the quantization matrix is implicitly solved without decoding the signal, and it is not necessary to make a determination, and the processing amount can be reduced.

  In the second embodiment, the quantization matrix determination unit 109d quantizes any one of the first quantization matrix and the second quantization matrix for each 4 × 4 transform block corresponding to the sequence for which the transform skip is enabled. The determination of whether to use is omitted. As a result, the amount of processing of the encoding process can be reduced and the encoding process can be speeded up.

(2.1.2) Operation of Encoding Process Next, an example of the encoding process according to the second embodiment will be described. FIG. 7 is a flowchart showing an example of the encoding process according to the second embodiment. The flow shown in FIG. 7 executes encoding processing for each transform block. Further, the flow shown in FIG. 7 is applied to the conversion block determined to be a 4 × 4 conversion block by the division control unit 109a. In addition, about the process same as the encoding process (refer FIG. 4) which concerns on 1st Embodiment mentioned above, the same code | symbol is attached | subjected and description is suitably used.

  As shown in FIG. 7, in step S301, the TS validation determination unit 109b determines whether to enable conversion skipping on a sequence basis (that is, whether or not the TS mode can be selected). If it is determined that the conversion skip is not activated in sequence units (step S301: NO), the same process as the conventional encoding process not using the conversion skip (TS mode) is performed in step S302. Thereafter, in step S106, the transform unit 104 performs orthogonal transform on the 4 × 4 transform block to be processed. Then, the process proceeds to S107.

  On the other hand, when it is determined that the conversion skip is enabled in sequence units (step S301: YES), the TS validation determination unit 109b is a control signal indicating that the conversion skip is enabled in sequence units (for example, transform_skip_enabled_flag = 1). Output). The quantization matrix determination unit 109d quantizes the quantization matrix including flat quantization values as a quantization matrix for the 4 × 4 conversion block based on the control signal indicating that the conversion skip is enabled in units of sequences. It is set in the unit 105 and the inverse quantization unit 106. For example, Equation (1) is applied as a flat quantization matrix. In this case, the quantization matrix determination unit 109d determines which of the first quantization matrix and the second quantization matrix is used for quantization for each 4 × 4 transform block corresponding to the sequence for which the transform skip is enabled. Omit the judgment. Thereafter, the process proceeds to step S103.

  In step S103, the TS application determination unit 109c determines whether to skip the conversion process (orthogonal transformation). The subsequent operation is the same as the encoding process (see FIG. 4) according to the first embodiment.

  Thus, for each 4 × 4 transform block corresponding to the sequence for which transform skip is enabled, the determination of which of the first quantization matrix and the second quantization matrix to use for quantization is omitted and flat. Use quantization matrix for quantization. As a result, high-speed, low-power consumption and high-quality encoding processing can be performed without causing image quality deterioration caused by applying the TS mode when setting a non-flat quantization matrix.

(2.2) Image Decoding Device An image decoding device 2 according to the second embodiment will be described.

(2.2.1) Configuration of Image Decoding Device The functional block configuration of the image decoding device 2 according to the second embodiment is the same as the functional block configuration (see FIG. 5) of the image decoding device 2 according to the first embodiment. is there. However, the operation of some of the blocks is different from that of the first embodiment.

  The image decoding apparatus 2 according to the second embodiment includes an inverse quantization unit 204 that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image; The inverse transform unit 205 which performs inverse orthogonal transform on the transformed block, and whether or not the transform skip for skipping the orthogonal transform is enabled for the 4 × 4 transform block having the 4 × 4 size, 1 or And a TS validation determination unit 200b that determines in units of a sequence of a plurality of frames. The inverse transform unit 205 is configured to generate a first quantization matrix including elements of different values for each 4 × 4 transform block corresponding to the sequence for which it is determined that the transform skip is enabled by the TS validation determination unit. Inverse quantization is performed using the second quantization matrix consisting of elements of equal value without using.

  In the second embodiment, the TS validation determination unit 200b determines whether the conversion skip is enabled in sequence units based on a control signal (for example, transform_skip_enabled_flag) indicating whether or not the conversion skip is enabled in sequence units. It is determined whether or not it is possible (that is, whether or not the TS mode can be selected), and a control signal indicating the determination result is output to the encoded data decoding unit 203.

  For example, when transform_skip_enabled_flag = 1, the TS enabled determination unit 200b determines that the conversion skip is enabled on a sequence basis. If transform_skip_enabled_flag = 0, the TS validation determination unit 200b determines that the transform skip is not enabled in units of sequences. The TS validation determination unit 200b may output a control signal indicating that a flat quantization matrix is to be used to the encoded data decoding unit 203, when it is determined that the conversion skip is enabled in units of sequences.

  In the second embodiment, the inverse quantization unit 204 is based on a control signal (or a control signal indicating that a flat quantization matrix is used) indicating that conversion skip is enabled on a frame basis or a sequence basis. , And for a transform block corresponding to a frame of a sequence for which transform skip is enabled, without using a first quantization matrix (non-flat quantization matrix) composed of elements of different values, Inverse quantization is performed using two quantization matrices (flat quantization matrices). In other words, the inverse quantization unit 204 always performs inverse quantization on the 4 × 4 transform block for the frame of the sequence determined by the TS validation determination unit 200 b to be the transform skip enabled. Use a second quantization matrix. This makes it possible to prevent the first quantization matrix from being applied to the 4 × 4 transform block from which the orthogonal transform has been skipped.

  In the second embodiment, the quantization matrix determination unit 200d inverts any one of the first quantization matrix and the second quantization matrix for each 4 × 4 transform block corresponding to the frame of the sequence for which the transform skip is enabled. The determination of whether to use for quantization is omitted. As a result, the processing amount of the decoding process can be reduced and the decoding process can be speeded up.

  The quantization matrix determination unit 200d (and the quantization matrix setting unit 206) relates to the quantization matrix included in the encoded data decoded by the lossless code decoding unit 202 for the frame of the sequence for which the conversion skip is enabled. Coded signals (such as candidate matrix data) may be ignored.

(2.2.2) Operation of Decoding Process Next, an example of the decoding process according to the second embodiment will be described. FIG. 8 is a flowchart showing an example of the decoding process according to the second embodiment. In the flow shown in FIG. 8, the decoding process is performed for each conversion block. Also, the flow shown in FIG. 8 is applied to the conversion block determined to be a 4 × 4 conversion block by the block size determination unit 200 a. In addition, about the process same as the decoding process (refer FIG. 6) which concerns on 1st Embodiment mentioned above, the same code | symbol is attached | subjected and description is suitably used.

  As shown in FIG. 8, in step S201, the encoded data decoding unit 203 receives a control signal and quantized transform coefficient data as a part of encoded data from the lossless code decoding unit 202, and proceeds to step S401.

  In step S401, the TS validation determination unit 200b determines whether the conversion skip is enabled in sequence units based on a control signal (for example, transform_skip_enabled_flag) indicating whether the conversion skip is enabled in sequence units. Determine When it is determined that the conversion skip is not activated in sequence units (step S401: NO), the same processing as the conventional decoding processing not using the conversion skip (TS mode) is performed in step S402.

  On the other hand, when it is determined that the conversion skip is enabled in sequence units (step S401: YES), the TS validation determination unit 200b codes a control signal indicating that a flat quantization matrix is used in the 4x4 block. It is output to the coded data decoding unit 203. Then, in step S203, the quantization matrix setting unit 206 sends candidate matrix data representing a predetermined flat quantization matrix to the inverse quantization unit 204 based on the control signal indicating that the flat quantization matrix is to be used. Set The quantization value indicated by the candidate matrix data may be, for example, m [x] [y] shown by equation (1). The subsequent operation is the same as the decoding process (see FIG. 6) according to the first embodiment.

  Thus, in the case where coding is performed by skipping orthogonal transformation for a coding sequence, it is implicitly determined that a flat quantization matrix is used in the 4 × 4 transform block. As a result, the process relating to the setting of the quantization matrix becomes unnecessary, and the process can be reduced. Also, since it is implicitly set, it is possible to perform normal decoding processing even if an error is superimposed on the coded signal related to the quantization matrix, and at the same time a virtual code with a small amount of code to be substituted as coding information. Code amount can be reduced by replacing with data.

(2.3) Summary of the Second Embodiment In the image coding device 1 according to the second embodiment, the quantization unit 105 performs 4 × 4 conversion in the sequence in which the conversion skip is enabled by the TS validation determination unit 109 b. Instead of using a first quantization matrix (non-flat quantization matrix) consisting of elements of different values for the block, using the second quantization matrix (flat quantization matrix) consisting of elements of equal values Perform. Further, in the image decoding device 2 according to the second embodiment, the inverse quantization unit 204 performs a first quantization including elements of different values with respect to the 4 × 4 transform block corresponding to the sequence for which the transform skip is enabled. Inverse quantization is performed using a second quantization matrix (flat quantization matrix) consisting of elements of equal values without using a matrix (non-flat quantization matrix). As a result, the first quantization matrix can be prevented from being applied to the 4 × 4 transform block in which the orthogonal transform is skipped, so that the image quality degradation due to the quantization distortion of the 4 × 4 transform block in the HEVC scheme It can be suppressed. In particular, the image quality of an image in which CG images or natural images and artificial images such as CG images are mixed can be improved.

  Furthermore, in the image coding device 1 according to the second embodiment, the quantization matrix determination unit 109d determines the first quantization matrix and the second quantization matrix for each 4 × 4 transform block corresponding to the sequence for which the transform skip is enabled. The determination of which of the quantization matrices to use for quantization is omitted. Furthermore, in the image decoding device 2 according to the second embodiment, the quantization matrix determination unit 200d determines the first quantization matrix and the second quantum for each 4 × 4 transform block corresponding to the sequence for which the transform skip is enabled. The determination of which of the quantization matrices to use for inverse quantization is omitted. By omitting this determination, the processing amount of the encoding process and the decoding process can be reduced, and the encoding process and the decoding process can be speeded up.

(3) Other Embodiments The first and second embodiments described above may be implemented as an image processing system including the image encoding device 1 and the image decoding device 2. The image processing system may include a storage medium for storing the encoded data output from the image encoding device 1, and the image decoding device 2 may receive the encoded data read from the storage medium. Further, the image processing system may include a network for transmitting the encoded data output from the image encoding device 1, and the image decoding device 2 may receive the encoded data transmitted from the network. The network may be wired or wireless, or may be part or all of a broadcast channel for transmitting data simultaneously to a plurality of transmission destinations.

  The conversion block is not limited to a square block in which the number of elements in the horizontal direction and the number of elements in the vertical direction are equal as described above. The conversion block may be a rectangular block in which the number of elements in the horizontal direction and the number of elements in the vertical direction are different, and is a linear block in which either the number of elements in the horizontal direction or the number of elements in the vertical direction is one. It is also good.

  Note that part of the image encoding device 1 or the image decoding device 2 in the first and second embodiments described above, for example, the subtraction unit 102, the conversion unit 104, the quantization unit 105, the inverse quantization unit 106, and the inverse conversion unit 107, quantization matrix setting unit 108, control unit 109, addition unit 110, deblocking filter 111, prediction unit 113, lossless encoding unit 114, control unit 200, lossless code decoding unit 202, dequantization unit 204, inverse transform The functions of the unit 205, the quantization matrix setting unit 206, the addition unit 207, the deblocking filter 208, and the prediction unit 210 may be realized by a computer. In that case, a program for realizing these functions may be recorded in a computer readable recording medium, and the computer system may read the program recorded in the recording medium and execute the program. Good. Here, the “computer system” is a computer system built in the image encoding device 1 or the image decoding device 2 and includes an OS and hardware such as peripheral devices. The term "computer-readable recording medium" refers to a storage medium such as a flexible disk, a magneto-optical disk, a ROM, a portable medium such as a ROM or a CD-ROM, or a hard disk built in a computer system. Furthermore, the “computer-readable recording medium” is one that holds a program dynamically for a short time, like a communication line in the case of transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case may also include one that holds a program for a certain period of time. The program may be for realizing a part of the functions described above, or may be realized in combination with the program already recorded in the computer system.

  Further, part or all of the image encoding device 1 and the image decoding device 2 in the first and second embodiments described above may be realized as a semiconductor integrated circuit such as an LSI (Large Scale Integration). Each functional block of the image encoding device 1 and the image decoding device 2 may be individually processorized, or part or all may be integrated and processed. Further, the method of circuit integration is not limited to LSI's, and implementation using dedicated circuitry or general purpose processors is also possible.

  As described above, the first and second embodiments have been described in detail with reference to the drawings, but the specific configuration is not limited to the above-described ones, and various design changes can be made without departing from the scope of the invention. It is possible.

1: image coding apparatus 101: frame buffer 102: subtraction unit 103: coded data generation unit 109: control unit 110: addition unit 111: deblocking filter 112: reference memory 113: prediction unit , 114: lossless encoding unit, 115: accumulation buffer, 116: motion vector calculation unit, 2: image decoding apparatus, 200: control unit, 201: accumulation buffer, 202: lossless code decoding unit, 203: encoded data decoding unit , 207: addition unit, 208: deblocking filter, 209: reference memory, 210: prediction unit, 211: frame buffer, 212: motion vector calculation unit

Claims (10)

A transform unit that performs orthogonal transform on a transform block that is part of a residual image that is a residual between the predicted image and the input image;
An enable determination unit that determines whether to enable the conversion skip for skipping the orthogonal transformation for a 4 × 4 transform block having a size of 4 × 4 in units of a sequence of one or more frames;
For each 4 × 4 transform block corresponding to the sequence determined to have the transform skip validated by the validation determination unit, the same value of the 4 × 4 transform block is used without using the first quantization matrix including elements of different values. An image coding apparatus comprising: a quantization unit that performs quantization using a second quantization matrix composed of elements. The apparatus further comprises a quantization matrix determination unit that determines which one of the first quantization matrix and the second quantization matrix is to be used for quantization of the 4 × 4 transform block in frame units,
The quantization matrix determination unit determines the first quantization matrix and the second quantization matrix for each 4 × 4 transform block corresponding to the sequence determined by the validation determination unit as having the conversion skip validated. The image coding apparatus according to claim 1, wherein the determination as to which one of them is used for quantization is omitted. An inverse quantization unit that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image;
An inverse transform unit that performs inverse orthogonal transform on the transform block subjected to the inverse quantization;
The 4 × 4 transformation block having the 4 × 4 size is provided with an validation determination unit that determines whether or not a transform skip for skipping orthogonal transformation is enabled in a sequence unit of one or a plurality of frames. ,
The inverse transform unit is a first quantization matrix including elements of different values with respect to each 4 × 4 transform block corresponding to the sequence determined to be enabled by the validation determination unit. An image decoding apparatus characterized by performing inverse quantization using a second quantization matrix consisting of elements of equal values without using. The apparatus further comprises a quantization matrix determination unit that determines, in frame units, which of the first quantization matrix and the second quantization matrix is to be used for inverse quantization with respect to the 4 × 4 transform block,
The quantization matrix determination unit determines the first quantization matrix and the second quantization matrix for each 4 × 4 transform block corresponding to the sequence determined by the validation determination unit as having the conversion skip validated. 4. The image decoding apparatus according to claim 3, wherein the determination as to which of the above is to be used for inverse quantization is omitted. A transform unit that performs orthogonal transform on a transform block that is part of a residual image that is a residual between the predicted image and the input image;
A quantization unit that performs quantization on the transform block;
For a 4 × 4 transform block having a size of 4 × 4, it is determined whether to use the first quantization matrix of different value elements or the second quantization matrix of equal value elements for quantization. And a quantization matrix determination unit that determines in units,
The conversion unit is configured to perform a conversion skip for skipping the orthogonal conversion for each 4 × 4 conversion block corresponding to a frame determined to use the first quantization matrix for quantization by the quantization matrix determination unit. An image coding apparatus characterized by performing the orthogonal transform on the 4 × 4 transform block without applying it. The method further comprises an application determination unit that determines whether to apply the conversion skip for each of the conversion blocks,
The application determining unit determines whether the conversion skip is applied to each 4 × 4 conversion block corresponding to a frame determined to use the first quantization matrix for quantization by the quantization matrix determining unit. The image coding apparatus according to claim 5, wherein the determination is omitted. An inverse quantization unit that performs inverse quantization on a transform block that is a part of a residual image that is a residual between a predicted image and an input image;
An inverse transform unit that performs inverse orthogonal transform on the transform block subjected to the inverse quantization;
For a 4 × 4 transform block having a size of 4 × 4, which of the first quantization matrix consisting of elements of different values and the second quantization matrix consisting of elements of equal values is to be used for inverse quantization in a frame unit And a quantization matrix determination unit to determine
The inverse transformation unit is a transformation that skips the inverse orthogonal transformation for each 4 × 4 transformation block corresponding to a frame determined to use the first quantization matrix for quantization by the quantization matrix determination unit. An image decoding apparatus characterized by performing inverse orthogonal transform on the 4 × 4 transform block without applying skips. The method further comprises an application determination unit that determines whether to apply the conversion skip for each of the conversion blocks,
Whether the application determining unit applies the conversion skip to each 4 × 4 transform block corresponding to a frame determined to use the first quantization matrix for inverse quantization by the quantization matrix determining unit The image decoding apparatus according to claim 7, wherein the determination of is omitted. On the computer
A transformation step of performing orthogonal transformation on a transformation block which is a part of a residual image which is a residual between a predicted image and an input image;
An enabling determination step of determining whether to enable conversion skipping for skipping the orthogonal transformation for a 4 × 4 transform block having a size of 4 × 4 in units of a sequence of one or more frames;
For each 4 × 4 transform block corresponding to the sequence determined to have the transform skip validated by the validation determination step, without using a first quantization matrix consisting of elements of different values, the same value And a quantization step of performing quantization using a second quantization matrix composed of elements. On the computer
An inverse quantization step for performing inverse quantization on a transform block which is a part of a residual image which is a residual between a predicted image and an input image;
Performing inverse orthogonal transformation on the transformation block subjected to the inverse quantization;
Execute an enabling determination step of determining whether or not conversion skipping for skipping orthogonal transformation is enabled for a 4 × 4 conversion block having a size of 4 × 4 in units of a sequence of one or more frames ,
The inverse transformation step is a first quantization matrix including elements of different values with respect to each 4 × 4 transformation block corresponding to the sequence determined to have the transformation skip enabled by the validation determination step. An image decoding program characterized by performing inverse quantization using a second quantization matrix consisting of elements of equal values without using.